Mr. Thompson illustrates the effect of the friction of the earth's surface on the eastward circulation of the air by a very simple experiment with a pail of water. If we put into the pail grains of any material a little heavier than water, and then give the water a rotatory motion by stirring it, the grains ought, by the centrifugal force imparted to them, to collect around the sides of the pail; but, sinking to the bottom, they do in fact tend to collect at the centre, carried inward by those currents which the friction of the sides and bottom indirectly produces. Thus Mr. Thompson's beautiful and philosophical theory completes that of Halley, and explains all those apparent anomalies which have hitherto seemed irreconcilable with the only rational account of the trade-winds. The rainless calms of the tropics are explained by this theory without that crossing and interference of winds which Lieut. Maury supposes; for the secondary circulation returns as an under-current toward the poles without reaching the tropics, and the dry lower current of the principal circulation passes over the tropical latitudes, in its gradual descent, before it reaches the earth as the trade-winds. These trade-winds, absorbing moisture from the sea, precipitate it as they rise again, and produce the constant equatorial rains; and these rains, doubtless, tend much more powerfully than the mere unequal distribution of heat to direct the wind toward the equator; for the fall of rain rapidly diminishes the pressure of the air and disturbs its equilibrium, so that violent winds are frequently observed to blow toward rainy districts. Thus, primarily, the unequal distribution of heat, and, more immediately, the equatorial rains cause the principal circulation of our atmosphere; and this indirectly produces the secondary circulation of Mr. Thompson's theory. Both these regular movements are, however, greatly disturbed, and especially the latter, by winds which are occasioned by local and irregular rains. In these movements and their causes we have the general outline of our subject, within which we must now sketch the weather. The causes of atmospheric movement, which we have thus far considered, are the unequal distribution of the sun's heat, the absorption and precipitation of moisture, the direct and the inductive action of the earth's rotation and friction. If to these we should add the tidal action of the sun's and moon's attractions, we should perhaps complete the list of _vera causae_ which are certainly known to exert a more or less general influence upon the atmosphere. But this short list is long enough, as we shall soon see.