By the annual and diurnal revolutions of the earth, each column of air is alternately exposed to the heat and cold of summer and winter, of day and night, and also to variations in the attraction of the sun and moon, which disturb its equilibrium, and produce tides similar to those in the ocean. Those produced by the moon ebb and flow twice during a lunation, and diurnal variations in the barometer, to a very small amount, are also due to the moon’s attraction.[134] The annual undulations occasioned by the sun have their greatest altitudes at the equinoxes, and their least at the solstices, and the diurnal variations in the height of the barometer, which accomplish their rise and fall twice in 24 hours, are chiefly due to the effects of temperature on the dry air and moisture of the atmosphere, which, according to Mr. Dove’s discoveries, produce independent pressures upon the mercurial column. A quantity of vapour is continually raised by the heat of the sun from the surface of the globe, which mixes in an invisible state with the dry air or gaseous part of the atmosphere. It is most abundant in the torrid zone, and, like the heat on which it depends, varies with the latitude, the season of the year, the time of the day, the elevation above the sea, and also with the nature of the soil, the land, and the water. There is no chemical combination between the aërial and aqueous atmospheres, they are merely mixed; and the diurnal variations arise from the superposition of two distinct diurnal oscillations, each going through its complete period in 24 hours; one taking place in the aërial atmosphere from the alternate heating and cooling of the air, which produce a flux and reflux over the point of observation; the other arising from the aqueous atmosphere, owing to the alternate production and destruction of vapour by the heat of the day and the cold of the night. The diurnal variations of the vapour have their maximum at or near the hottest hour of the day, and their minimum at or near the coldest, which is exactly the converse of the diurnal variations of the dry air. On the whole, there are two maxima and two minima heights of the barometer in the course of 24 hours from the combinations of these, but in the interior of continents far from water, where the air is very dry, there ought to be one maximum and one minimum during that period, according to this theory.