down its load of moisture, only a portion of it can Lo taken up again; the rest is absorbed by the earth to feed the springs. On the polar side of 40° S. we have a water instead of a land surface, and as fast as precipitation takes place there, the ocean replenishes the air with moisture again. It may consequently be assumed that a high dew-point,—at least one as high as the ocean can maintain in contact with winds blowing over it, and going from warmer to cooler latitudes all the time—is the normal condition of the air on the polar side of 40° S., whereas on the polar side of 40° N. a low dew-point prevails. The rivers to the north of 40°, I reckon, could not, if they were all converted into steam, supply vapour enough to make up this average difference of dew-point between the two hemispheres. The symmetry of the rain and storm curves on the polar side of 40° S. suggests that it is the condensation of this vapour which, with the liberation of its latent heat, gives such activity and regularity to the circulation of the atmosphere in the other hemisphere.
827. The rain-fall of Cape Horn and Cherraponjie.—On the polar side of 40° S., near Cape Horn, the gauge of Captains King and Fitzroy showed a rain-fall of 153.75 inches in 41 days. There is no other place except Cherraponjie where the precipitation approaches this in amount. Cherraponjie (§ 299) is, it has already been stated, a mountain station in India, 4500 feet high, which, in latitude 25° N., acts as a condenser for the monsoons fresh from the sea. But on the polar side of latitude 45°, in the northern hemisphere, it is, except along the American shores of the North Pacific, a physical impossibility that there should be a region of such precipitation as King and Fitzroy found on the western slopes of Patagonia—a physical impossibility, because that peculiar combination of conditions required to produce a Patagonian rain-fall is wanting on the polar side of 45° N. There is in the North Atlantic, water surface enough to afford vapour for such an amount of precipitation. In the North Pacific the water surface may be broad and ample enough to afford the vapour, but in neither of these two northern sheets of water are the winds continuous enough from the westward to bring in the requisite quantities of vapour from the sea. Moreover, if the westerly winds of the extra-tropical north were as steady and as strong as are those of the south, there is lacking in the north that continental relief—mountain ranges rising ab-
