THE SUN — ITS CHEMICAL ANALYSIS. 187 



it is hard to comprehend why all the substances it contains have not 

 been gradually raised, however feeble the conductibility, to the tem- 

 perature of incandescence. One may keep for a time a piece of ice 

 in a warm chamber, but it always ends by melting. Suppose the 

 nucleus of the sun as cold as you please, sufficiently cold to be inhab- 

 itable by beings like ourselves, the radiation of that furnace which we 

 call the photosphere must gradually elevate the temperature of the 

 first atmosphere and that of the solid globe itself. I think, therefore, 

 that M. Kirchoff has reason to reject a theory which is opposed to 

 all the known laws of the diffusion of heat. 



There still remains, however, the celebrated observation of Arago. 

 If, as M. Kirchoff contends, the luminous disc is not a gaseous photo- 

 sphere, how is it that it transmits to us unpolarized light? On this 

 point, the German savant remarks that if liquid luminous bodies emit 

 polarized light, it is because we observe them when at rest; if we 

 examined them when agitated, he thinks that the light being then 

 emitted under the most different angles there would be no uniformity 

 in the direction of the vibratory movements, consequently no polar- 

 ization. He conceives, therefore, that the visible surface of the sun 

 may very well be liquid and yet emit unpolarized or natural light, 

 because that vast ocean of tire has not a smooth surface like a mirror, 

 but is unceasingly furrowed by enormous w^aves and swept by tremen- 

 dous tempests. To form such suppositions is not to yield to the sway 

 of the imagination. All movement of air or water results from a 

 simple difference of temperature : some degrees less from the pole to 

 the equator on our earth, and we see the ocean traversed by currents 

 and counter-currents, the atmosphere open to the winds, agitated by 

 tempests; but the fluctuations of temperature, so comparatively slight 

 on our planet, must, if we reflect, be immense in the sun and around 

 the sun. The condensation of the metallic vapors of the atmosphere, 

 the vaporization from the solar ocean, are phenomena far otherwise 

 stupendous than our terrestrial thunder-storms. What force must 

 the winds have on a sphere so vast as that of the sun ! What fear- 

 ful deluges must pour from the bosom of clouds as extensive as our 

 widest continents and charged with metals in ignition ! The varia- 

 tions of temperature in the solar atmosphere may be estimated, with 

 great chances of probabihty, at many hundreds of degrees; the atmos- 

 pheric pressure must vary in the same proportions, while it is by 

 millimetres only that, in terrestrial barometers, we compute the varia- 

 tions of pressure in the mercurial column which forms the equilibrium 

 to the weight of the atmosphere. And yet, we know a rapid fall of 

 some millimetres in the barometric column is the certain forerunner 

 of a violent tempest. 



After having described the sun as an incandescent liquid globe, 

 encompassed with a dense atmosphere pervaded by simple bodies, 

 which we find in our own planet, M. Kirchoff stops. He seeks not 

 to explain how this focus of heat has been kindled, nor how it is 

 maintained. This question, if it cannot be completely resolved, de- 

 serves, at least, to be discussed. If the sun were only a heated body 

 radiating in space, like a red-hot cannon ball, we must conclude that 



