In 1760 appeared the very remarkable Paper, in the fifty-first volume of the "Philosophical Transactions," of the Rev. John Mitchell, of Cambridge, in which he views an Earthquake as a sudden lifting up, by a rapid evolution of steam or gas beneath, of a portion of the earth's crust, and the lateral transfer of this gaseous bubble beneath the earth's crust, bent to follow its shape and motion, or that of a wave of liquid rock beneath, like a carpet shaken on air. Great as are certain collateral merits of Mitchell's Paper, showing observation of various sorts much in advance of his time, this notion of an Earthquake is such as, had he applied to it even the imperfect knowledge of mechanics and physics then possessed in a definite manner, he could scarcely have failed to see its untenable nature. That the same notion, and in a far more extravagant form, should have been reproduced in 1843 by Messrs. Rogers, by whom the gigantic parallel anticlinals, flanks and valleys of the whole Appalachian chain of mountains are taken for nothing more than the indurated foldings and wrinkles of Mitchell's carpet, is one of the most salient examples of the abuse of hypothesis untested by exact science. Neither Humboldt nor Darwin, great as were the opportunities of observation enjoyed by both, can be supposed to have formed any definite idea of _what_ an Earthquake is; and the latter, who had observed well the effects of great sea-waves rolling in-shore after the shock, did not establish any clear relation between the two.[A] Hitherto no one appears to have formed any clear notion as to what an Earthquake is--that is to say, any clear idea of what is the nature of the movement constituting the shock, no matter what may be the nature or origin of the movement itself. The first glimmering of such an idea, so far as my reading has enabled me to ascertain, is due to the penetrating genius of Dr. Thomas Young, who, in his "Lectures on Natural Philosophy," published in 1807, casually suggests the probability that earthquake motions are vibratory, and are analogous to those of sound.[B] This was rendered somewhat more definite by Gay Lussac, who, in an able paper "On the Chemical Theories of Volcanoes," in the twenty-second volume of the "Annales de Chémie," in 1823, says: "En un mot, les tremblements de terre ne sont que la propagation d'une commotion à travers la masse de la terre, tellement indépendante des cavités souterraines qu'elle s'entendrait, d'autant plus loin que la terre serait plus homogène."