We know the quantity of heat which each substance throws off in a state of combustion; we know, too, what a vast body the sun is; and we are able to calculate with a rough but sufficient approximation the quantity of heat which the body of the sun would produce in burning. The result of this calculation is, that, at the elevated temperature which the sun possesses, the combustion of the solar mass could not be kept up during many ages. Since the historic period this temperature would have been so lowered as to produce a change in the seasons that has not taken place. We are compelled, then, to abandon the idea of a mass in combustion, as well as that of a luminous globe, and to acknowledge that there is a secret which has escaped us.

This secret, gentlemen, chemistry is charged to unveil to us. Astronomers profit eagerly by all the discoveries which physical science makes, and it is by this means alone that they arrive first at conjecture, and afterward at a knowledge of what is taking place at prodigious distances. It is thus that the phenomenon of dissociation recently discovered by M. Sainte-Claire Deville, puts us in the way of explaining the permanence of the solar temperature. We know that no combination can resist heat. Whatever may be the stability of the combination, whatever energy the affinitive force may possess, if the temperature is raised to the proper degree, the elements separate, and remain together simply in a mixed state, wanting to combine anew when the temperature is lowered. This is what we call dissociation; and this is just the state, for example, in which we find oxygen and hydrogen gas, exposed to a temperature of 2500 degrees. At such a temperature they remain in a mixed state, without being able to form water, which ought to result from the combination of these two elements. But the phenomenon of dissociation cannot take place without the intervention of an enormous amount of heat. To illustrate this, let us suppose a kilogram of ice at zero. In liquefying it would absorb 79 degrees of heat; to make it warm, 100 degrees would be required; in evaporation it would absorb 640; and to dissociate it, 3955, or nearly 4000 degrees would be necessary. What we say of water is equally true of all the combinations; all that is required being to change the numerical degrees of the latent heat, for fusion, for volatilization, and for dissociation. This being so, we arrive at the conclusion that even the least considerable quantity of matter in a state of dissociation may be regarded as a magazine of latent heat continually tending toward sensible development.

The temperature of dissociation of water is almost 2500 degrees. The temperature of the sun being at least five millions of degrees, the whole mass of which it is composed ought to be in a state of dissociation, and to contain consequently an enormous quantity of latent heat independent of the sensible heat; to which is owing this prodigiously elevated temperature. What, then, is the effect which the solar matter ought to produce on the radiation of which it is the seat? Almost the same effect that radiation produces on a liquid body which has reached a temperature of solidification. The heat necessary to keep up the radiation is borrowed from that part of the liquid which solidifies, so that the temperature, instead of decreasing, remains constantly at the point at which solidification ceases. This is really what passes on the surface of the sun. This brilliant mass, raised to a temperature of five millions of degrees, has a tendency to cool itself rapidly. The radiation produces, in fact, a coolness in the superficial stratum. By reason of this coolness, part of the gas which composes the atmosphere is lowered below the temperature of dissociation; it yields then an enormous quantity of heat, which from latent becomes sensible, and prevents also an ulterior lowering of temperature. It is sufficient to repair the continual loss of heat that a mass of several kilograms passes daily from a state of dissociation to one of combination; and it is evident, considering the enormous size of the solar globe, that things may remain in this state during millions of ages without the temperature of the sun changing in a manner which may be felt by us. I say, by us, for our knowledge of this temperature is obtained at no less a distance than several hundred thousands of degrees.

It appears, then, from the very nature of the sun, that it does not possess an inexhaustible quantity of latent heat. A day will come when it will no more be able to lose heat without being cooled in a sensible manner, but that cooling will not take place before a very distant period, and long after we have disappeared from this world.

By way of recapitulation of the several views we have set forth, let us endeavor to give you a precise idea of the sun, as regards both its interior and its surface. The reasonings which we have just advanced, founded partly on astronomical observations and partly on known principles of science, lead us to regard the sun as composed of a fluid or gauze-like mass, surrounded with a photospheric stratum, the matter of which has passed through the first stage of condensation. According to the views held by Laplace, the sun proceeded from the hands of its creator in a nebulous state. We are led to believe that the interior mass is still in this state. A change has taken place only on the surface, because there only could the loss of heat owing to radiation produce a partial cooling. The result of this cooling is the condensation of a relatively small quantity of matter, which, possessing a very considerable power of emission, forms the photosphere. It is in the presence of this photosphere that the only difference exists between the sun and a nebula, between the myriads of stars which people the heavens, and the nebulae with whose existence the telescope makes us acquainted.

We come, at length, to the last with which we proposed to deal: What is the constituent matter of the sun? What are the elements which enter into the composition of its atmosphere and of the photospheric bed? Some years ago, to put a question like this would have been regarded as rashness; to attempt to answer it, the height of folly. We only knew, from the analysis of meteoric stones, that cosmical matter did not contain any other elements besides those of which our globe is composed. But to-day we can go further, thanks to the discoveries of the German Kirchoff.

We all know the solar phantom, and the brilliant colors which result from the decomposition of the white light. This phantom seems continuous if we make the observations in a rough manner; but if we employ delicate means, we see that it is formed of a multitude of black streaks and of brilliant rays perfectly distinct from each other. It is impossible to imitate this appearance artificially. All that we are able to do is to project on a screen the figure of a solar appearance taken from a drawing. You see that it is furrowed over with a considerable number of black streaks, of which the principal ones are, according to Fraunhofer, who discovered them, indicated by the letters of the alphabet, A, B, C, etc. These streaks are extremely numerous: we have counted no fewer than 45,000 of them.

I have said that it is impossible for us to imitate this appearance with our artificial lights, and it is precisely here that we are able to discern the nature of the different sources of light. In fact, each source has an appearance peculiar to itself, and by which it is characterized. The brilliant line of the Drummond light gives a continuous appearance, and it is the same with all the simple incandescents. But when we analyze the light of a body in combustion, we arrive at an entirely different result. The appearance obtained in this case is crossed by rays which, instead of being black, are, on the contrary, more brilliant than the colors in the midst of which they are formed. The same thing happens when we make the rays emanating from the electric light pass through a prism, because in this case there is combustion, that is to say, a combination of the oxygen in charcoals, mixed with foreign matter, from which is produced the voltaic bow. If we are content to restore these burning coals, they will give a continuous appearance just as lime.

The brilliant spectral rays are not always the same. They depend on the nature of the metal which is found in the flame, and which takes part in the combustion. You see at this moment the appearance which silver presents: it is characterized by a magnificent green ray. Here is now the appearance of copper, which, we know, has a yellow ray, accompanied by a fine group of green rays, different from those which silver produces. We now burn some zinc, which gives a magnificent group of blue rays, a fine red ray, and another of violet. Finally, we shall close these experiments with burning brass, which is, as you are aware, a mixture of copper and zinc. You will recognize in the appearance which is produced the characteristic rays of those metals, each of them producing its proper effect, as if it were alone.