of verity." This also explains why the duration of the sun's rotation, twenty-five clays, is less than that of the revolution of the various planets. And the triple ring of Saturn seems to be an ocular proof of the original extension of the atmosphere of that planet, and of its successive contractions. So many analogous phenomena certainly render Laplace's cosmogonic hypothesis highly probable.
A final confirmation of the theory is supplied by spectrum analysis. The study of the spectra of the nebulae has demonstrated that, if many of them are merely agglomerations of stars, others are still gaseous bodies—veritable specimens of the primitive chaos, exemplifying perfectly Kant's, W. Herschel's, and Laplace's conception of the beginnings of worlds as they left the Creator's hands. Of the nebulæ two appear to be composed of a central globe with a ring like Saturn's, and in many others it seems possible to discern the gyratory movement by means of which planetary systems are formed.
Of recent investigations that have served to establish the basis and develop the results of Laplace's theory, we must place in the first rank the valuable researches of M. Edouard Roche, on the form of the heavenly bodies, which the author has recently supplemented by an essay on the constitution and origin of the solar system. M. Roche first demonstrates that by virtue of the particular form of the "free surface" bordering the atmosphere—a surface having a projecting ridge at the equator—as the nebula contracts a fluid stratum will slide from the poles toward the equator and be thrown off over the equatorial ridge as through an opening. It is thus that an equatorial zone, independent of the central body, will be formed and become an outer ring.
But the theory shows that inner rings will be formed from portions of the mobile matter brought toward the equator from the poles, and it is thus that Saturn's two inner rings would be formed, their radius being less than twice that of the planet. The equatorial extent of the planet's atmosphere being at present equal to 2, there can not have been a ring thrown off inside this distance. Laplace's theory, not admitting inner rings, accounts only for the formation of the largest of the three. M. Roche also holds that the moon was formed from an inner ring, and that it was developed in the bosom of the earth's atmosphere, which, withdrawing little by little, left its satellite free.
Every conception that favors Laplace's theory clearly tends to confirm the hypothesis of the earth's original fluidity, but without settling the question of the liquidity of the nucleus at the present time. Let us see to what extent this obscure question has been elucidated.
The equatorial swelling, which changes so slightly the spherical form of the globe, has nevertheless a very appreciable effect on the globe's rotation on its axis. If the earth were an exact sphere and entirely homogeneous, or if it were composed of homogeneous concentric spheres, the sun's attraction would have no effect on the move-
