The action of volcanoes is still more limited, and more local, than any of those which have yet been mentioned. Although we have no precise idea of the means by which nature keeps up these violent fires at such great depths, we can judge decidedly, by their effects, of the changes which they may have produced at the surface of the globe. After a volcano has announced itself, by some shocks of an earthquake, it forms for itself an opening. Stones and ashes are thrown to a great distance, and lava is vomited forth. The more fluid part of the lava flows in long streams, while the less fluid portion stops at the edges of the opening, raises its margins all round, and forms a cone, terminated by a crater. Thus volcanoes accumulate upon the surface matters which were previously buried in the bowels of the earth, after modifying their nature, and raise themselves into mountains. By these means, they have formerly covered some parts of our continent, and have also suddenly produced islands in the middle of the sea. But these mountains and islands have always been composed of lava, and all their materials have undergone the action of fire: they are disposed as matters should be, which have flowed from an elevated point. Volcanoes, therefore, neither raise nor overturn the strata through which their apertures pass; and if some causes acting from those depths have contributed, in certain cases, to raise up large mountains, they cannot have been volcanic agents of the same nature as those which exist at the present day. * * * * * Thus, we repeat, it is in vain that we search, among the powers which now act at the surface of the earth, for causes sufficient to produce the revolutions and catastrophes, the traces of which are exhibited by its crust: And if we have recourse to the constant external forces with which we are as yet acquainted, we shall have no greater success. _Constant Astronomical Causes._ The pole of the earth moves in a circle around the pole of the ecliptic, and its axis is more or less inclined to the plane of the ecliptic; but these two motions, the causes of which are now ascertained, are much too limited for the production of effects like those whose magnitude we have just been stating. At any rate, their excessive slowness would render them altogether inadequate to account for catastrophes which, as we have shewn, must have been sudden. The same reasoning applies to all other slow motions which have been conceived as causes of the revolutions in question, chosen doubtless in the hope that their existence could not be denied, because it might always be easy to hold out that their very slowness rendered them imperceptible. But whether they be true or not is of little importance, for they explain nothing, as no cause acting slowly could have produced sudden effects.