The last lecture but one of the series of admirable scientific expositions before the American Institute, was delivered last evening in Steinway Hall, by Professor Cook, of Harvard University, on the " Spectroscope." The hall was densely crowded. The Professor said: The color of light depends, to a certain limited extent, on the nature of the source from whence it proceeds, and by a study of the relation between these two we have reached a new method of chemical analysis, by wliich we have been enabled, not only to discover several new metallic elements among the materials of the globe, but also to extend our investigations beyond the limits of our globe, and to reach some knowledge, however indistinct as yet, respecting the sun, the planets, and the other heavenly bodies. To the course of reasoning through which these remarkable results have been attained, I am to ask your attention in my lecture this evening. All bodies when heated to a sufficiently high temperature emit light. Indeed, it can be readily shown that light is a necessary result of a high temperature. A bar of iron heated in a blacksmith's forge, a mass of coal burning in a grate, the gas which illuminates our hall, are all illustrations of this general principle. In all these cases the light emitted has no particular color, and is what we call a " puro white light." And the source from which light is emitted in every case is a solid body. Moreover, what is true in these limited instances, we find to be a universal truth, namely, that an incandescent solid body always emits pure white light. But before we can deduce any safe conclusion from this general principle, we must understand what is meant by pure white light—for white light is not, as is frequently supposed, a simple sensation—but, on the contrary, is a very complex sensation. It is to Sir Isaac Newton that we owe the first analysis of white light. He found, on passing a beam of white light through a glass prism like this that it became divided into different colored rays. We shall repeat his experiment late in the evening ; but for the moment the diagram I have here will enable me to explain the process he employed. If a beam of sunlight enters a dark room througlx a slit in a shutter, as any one knows, the beam crosses tlio room in a rectilinear direction, 'and forms on the opposite walls an image of the slit. But if wo interpose between the slit and the wall a glass lens, we shall obtain a perfectly different image of the slit, such as is indicated here. If now still further we interpose in the path of the same beam a prism, we find that the prism produces two distinct results. In the first place it bends the beam, so that the image of the slit, instead of being opposite upon the wall, is thrown very mucli to the left, and will fall very much in this direction. But in the second place, it spreads out the beam just like the rays of a fan ; and we have formed, instead of the single image of the slit, a broad band of blending colors called the " solar spectrum." If, now, by any means we recombine those different- 1 colored lights, we obtain — as I shall show you —the pure white light again. From all this it follows that the white light is a very complex sensation ; it is simply the confused impression produced upon the eye by the simultaneous effect of light of every different shade of color. Pure color, on the other hand, is a simple sensation; but most natural colors are not simple colors. Thus, for example, the color purple is a complex color, formed by the blending of tints of red and violet. Now, we can easily discover the different hues of which any given color consists, by simply passing the light that emanates from the source through a glass prism, when we shall divide it up into its simple constituent tints. On account of the great interest which attaches to studies of this kind, a class of instruments has been invented for analyzing colors, to which we give in general the name of " Spectroscope." We have one of them here before us. It is not necessary for me-to enter into any detailed account of the mechanism of this instrument, for it is exceedingly simple, as might be gathered from what has been said on the subject. It consists of several parts. We have here the slit through which the light enters; we have the glass prism which bends the ray, so that it passes down this second tube. Moreover, we have a small telescope to give a well defined image of the slit. If, then, we direct this instrument towards any source which gives us a pure colorless flame, we see simply an image of the slit; but if we direct it towards a colored light, or a light containing several simple hues, we see just as many distinct images of the slit as there are separate colors. And the reason of the diversity is this: the glass prism bends each separate color to a different extent. It bends the red color the least, and the Violet the most; and, therefore, those separate tints become sepaiated along the band which we call the spectrum. Now, if our colored light consists of one single tint, we get a single image. If it consists of three tints, we get three images. If it is pure j white light, which contains every gradation of color, then we have an infinite number of images of the slit, which succeed and overlap each other and thus blending form the band which I have already pointed out. Understanding now what is meant by "white light" we may return to the point from which we started—that an incandescent solid body always emits pure white light. What is true of ' solid bodies is also true of liquid bodies which can be heated 116 to a temperature where they become incandescent metals—as, for example, molten metals— they also emit white light. Mark, now, the important conclusion to which this necessarily leads: If we can reason from analogy it follows that whenever we see white light the luminous source is a solid or a liquid tody. And, so far as our experience of the surface of the earth goes, this is universally true. And analogy must lead us to suppose that it is a general law of nature. Now, the light which comes from the sun and moon and most of the fixed stars is a pure white light; then it would follow that the sun and the fixed stars are solid or liquid incandescent todies. The learned professor then proceeded to describe " absorption bands," especially those seen in the solar spectrum ; and concluded by exhibiting upon an immense screen upon the platform the spectra of all ths most important metals. These, of course, were extremely beautiful, and elicited unbounded applause from the audience. Value of Zinc for Roofing A correspondent asks, "What is the relative value of zinc and tin for roofing 1" The relative value of zinc and tin for roofing purposes depends so much upon variable conditions that it can hardly be stated in a simple affirmative or negative answer. Its comparative cost with other roofing material, slate, shingles, etc., cannot for the same reason be definitely stated. The gases arising from the combustion of bituminous coal, would not, in our opinion, under ordinary circumstances affect its durability. The connection of copper with zinc roofing is not advisable, as the galvanic action likely to occur under such an arrangement would render the zinc less permanent. A coat of any oil paint would tend, of course, to preserve it from the oxidization which takes place on its exposure to damp air. The coating of oxide which forma upon the surface of zinc thus exposed, tends, however, to protect the metal from further oxidization, sa that it will last a long time without any other protectioa. There is no doubt that for many localities it is one one of the best as well as cheapest of roofing materials.