METEOROLOGY It is true that in the interior of North America and of Asia the air at the surface is rarely if ever completely saturated; but on the other hand, the monthly and annual averages show that a very large amount of aqueous vapor is always present even in those regions, notwith- standing their comparative dryness. The dis- tribution of moisture in reference to altitude above the sea was for a long time supposed to follow the law suggested by Dalton (1806), namely, that it was expanded throughout the atmosphere precisely as if constituting an in- dependent vaporous atmosphere within the gaseous one ; but the later researches of Re- gnault, Bessel, Strachey (1861), and Hann (1874) have shown the fallacy of -this theory. According to Hann, the pressure diminishes as we ascend more rapidly than it should accord- ing to Dalton's law, and the weight of the vapor existing in a vertical column above a given place is only 0-22 of what his law would indicate ; five tenths of the aqueous vapor is within 6,500 ft. of the surface of the sea, and the strata under 20,000 ft. in altitude contain nine tenths of the vapor in the entire atmos- phere. The ratio between the quantity of vapor at a given height and that at the surface is shown in the following table, as deduced by Hann from balloon and mountain observations: Altitude, feet. Quantity of vapor. Altitude, feet. Quantity of vapor.
1,000 2,000 8,000 4,000 5,000 6,000 7,000 1-00 0-87 0-80 0-78 0-64 0-56 0-56 0-48 8,000 10,000 12,000 14,000 16,000 18,000 20,000 22,000 0-42 0-84 0-27 0-23 0-18 0-16 0-13 0-07 On the other hand, the irregularities of the dis- tribution of vapor are such that the observations made in England in balloons show that layers of moist and dry air may alternate with each other. In the United States, and generally in continental situations, it is probable that such alternations only attend storms. The relation between moisture and the direction of the wind is expressed graphically by the so-called hygro- metnc wind rose, which gives for each direc- tion of the wind the average force of vapor and relative humidity. The following table shows these relations for some of the few stations for which the computations have been made : nygrometric Wind Rose. FORCE OF VAPOR. RELATIVE HUMIDITY. London. Halle. Winter. Summer. Winter. Summer. N In. 0-21 in. 0-42 per ct. 68 N.E.. 0-20 0-41 91 67 E 8. E.. 0-19 0-27 0-49 0-64 98 86 61 66 8 0-82 0-60 88 67 8.W.. 0-82 0-54 82 70 W.... 0-28 0-49 81 71 N. W. 0-24 0-44 88 69 N.... 0-21 0-42 80 68 The relative dampness of the easterly winds and the dryness of the westerly winds is well shown by this table for Europe. We also see that the fact that the summer winds are dryer than the winter is due to their increased tem- perature and capacity for vapor, and not to any diminution in the quantity of vapor held by them, since the latter is in summer larger than in winter. 5. Precipitation of Aqueous Vapor. This is apparent under the forms of cloud, fog, dew, and rain, and their various modifica- tions, and produces a local diminution in the barometric pressure. The quantity of dew admits but very rarely of precise measurement, the most satisfactory series of experiments hav- ing been made by Dr. W. C. Wells in London. (For many details concerning it, see DEW, and FEOST.) However important an abundant dew may be to the husbandman, it can scarcely be considered by the meteorologist as other than a local and temporary phenomenon, of minor importance in the general economy of the atmosphere. The formation of fog is but little more important in this connection. (See FOG.) Of more importance to all classes must be esteemed the formation of clouds and rain. (See CLOUDS, and RAIN.) Clouds, ac- cording to the commonly received classifica- tion of Howard (1802), are divided into the cirrus, stratus, cumulus, and nimbus, for which latter Poey (1872) has proposed to substitute the pallium or sheet cloud. Howard intended by the term nimbus to designate those clouds from which rain was falling ; but it does not appear certain that rain may not occasionally fall from either the stratus, the cumulus, or the pallium, or from the numerous combina- tions of these typical forms. The presence of clouds and the phenomena displayed by them are valuable to the meteorological observer, both as affording him an indication of the di- rection of the upper currents of air, and as giving some clue to the moisture and tempera- ture there prevailing. For the present we have principally to do only with the average percentage of cloudiness, and its influence upon the temperature. A diurnal periodicity, show- ing two maxima and two minima in the per- centage of cloudy sky, will be perceived by the most cursory observer ; the hours at which these occur vary with the kind of cloud ; for the cumulus they are, maximum at 3 P. M., minimum at 5 A. M. ; for the stratus, maximum at 6 A. M., minimum at 3 P. M. In moun- tainous countries it is easy to trace the direct dependence of this periodicity upon the heat- ing of the soil during the day time and the cooling of the air by radiation during the night time; but the daily period is also distinctly pronounced within the tropics, and is clearly seen whenever a station in the temperate zone is in the interior of the continent within the range of an area of relatively dry air. The total percentage of all kinds of clouds is shown for two European stations in the following table for January and July :
