at the bottom, with cooler water or ice floating above it. However much the surface be cooled, water colder than 40° cannot descend to displace water warmer than itself. Hence we never can have ice formed at the bottom of deep water, though it is not uncommon to find it thus situated, in shallow streams or rivers of rapid flow. Here the temperature of the whole body of water is brought down to the freezing point, and in freezing the ice adheres to the sides and bottom of the stream. What a beautiful provision is this, that the coldest water should rise to the surface, and there freeze and remain, exposed to the warmth of the sun-beams and the air, to be speedily dissolved upon the return of spring! This is owing to the well known fact, that in the act of freezing a still further expansion takes place, so that the specific gravity of ice is less than water of any temperature, and consequently floats upon the surface. We thus see that by the contraction of water by cold, the temperature of various times and places is equalized, though were that contraction without limit, a great portion of the earth would be bound in fetters of ice. Such a disastrous result, is prevented by the substitution of expansion for contraction, when the temperature is reduced to 40°, and the benevolent purposes of an all-wise Designer, are made still more manifest by the further expansion of water in the act of freezing. As water becomes ice by cold, it becomes _steam_ by heat. We generally understand by steam the vapor of hot water, but steam or vapor rises from water at all temperatures, however low, and even from ice. The expansive force of this vapor increases rapidly as the heat increases, but yet in all cases the surface of water is covered with an atmosphere of aqueous vapor, the pressure, or _tension_ of which is limited by the temperature of the water. If, therefore, the vapor is not confined, causing the surface of water to be pressed upon, evaporation will take place, and thus there must, according to this law, always exist an atmosphere of aqueous vapor, the tension of which may be compared with that of our common atmosphere. Now the pressure of the latter is measured by the barometrical column, about 30 inches of mercury, while that of watery vapor is equal to one inch of mercury at the constituent temperature of 80 degrees, and to one fifth of an inch at the temperature of 32 degrees. If the atmosphere of air by which we are supported were annihilated, there would still remain, an atmosphere of aqueous vapor, arising from the waters and moist parts of the earth, but in the existing state of things this vapor rises _in_ the atmosphere of dry air, and thus its distribution and effects are materially influenced by the vehicle in which it is thus carried.