Vertical Deorease of Temperature.— From the Paris records the mean fall in temperature per 1,000 feet up to a height of 20,000 feet is Z4 in winter, 2.8 in spring, 2.6 in summer, 2.5 in autumn and 2.6 for the year. From the Berlin records it is 3.1 winter and spring and 3.0 in summer and autumn. It should be noted that the decrease is greater at Berlin, which is farther from the influence of the ocean than is Paris. In the interior of the United States it is still greater.
Isothermal Layer.— During the past 15 years the atmosphere has been explored to altitudes of from 15 to 30 kilometers by the sending up of sounding balloons equipped with suitable •instruments. We find that there are two distinct strata in our atmosphere that inter mingle but slightly: a lower turbulent one, in which all of the ascending and the descending whirls that constitute storms and cold waves operate, and in which the temperature decreases rapidly with elevation; and an upper or outer one, in which there is such small change of temperature with elevation, so far as observa tions have reached, that it is called the isother mal layer. The lower stratum is about 12 kilo meters in depth; the upper one extends to limits unknown. The average temperature for summer at the top of the storm stratum is — 52° C., and 8 kilometers higher in the iso thermal stratum, —51°. In winter the tem peratures at these two places are —58° and — 57° C. Note that the temperature slightly increases with elevation in the region above storms; although we feel justified in assuming that farther out the temperature must lose with altitude until at the outer limits of the atmos phere it approaches absolute zero. The greatest altitudes reached showed about one-half the difference in the temperature between winter and summer that is found at the earth's surface.
Diurnal Range of Temperature.— The average difference between the temperatures of day and those of night diminishes with ascent in the free air until at only a few thousand feet from the surface of the earth (3,000 to 5,000) there is no difference between the heat of midday and that of midnight. The diurnal range of temperature is greatest at the equator and diminishes irregularly with latitude, being greater over land than over water on the same parallel.
Annual Range of Temperature.— The an nual range of temperature increases with lati tude, being the greatest for the interior of continents and least for islands and near the oceans. It decreases with gain in elevation
in free air, being about 7° C. at 18 kilometers about the earth.
Frosts.— As previously explained, a given quantity of heat being absorbed by different substances of unequal specific heats will pro duce a different temperature in each; the lower the specific heat of a substance the higher will be its temperature; and besides different sub stances have different coefficients of absorption and reflection. Therefore ,under the same inso lation rocks, clay, vegetation and other sub stances come to temperatures that differ by many degrees. The air next to earth largely partakes of the temperature of the surface, and in consequence when the sky is clear so that incoming and outgoing radiations may progress freely and the air is little disturbed by wind, wide variations in the temperiture of a thin stratum of lower air may occur over adjacent plots of ground of precisely the same elevation, but different covering, and frost form. Differ ence in elevation may also cause marked differ ences in minimum air temperatures where the surface covering is uniform and frosts of vary ing degrees of intensity occur. During clear, cool nights with light winds, frosts will occur in bogs or on low, damp ground, while in the surrounding uplands the temperature may be 10° to 20° higher.
Atmospheric Pressure.— The attraction of gravity on the gases of the atmosphere causes them to exert a pressure on the surface of the earth. The average pressure at sea-level is about 14.7 pounds per square inch. If we re move the air from the inside of a vessel k means of the air pump or otherwise, then external pressure becomes at once apparc. This pressure is in all directions, upward a well as downward; it is measured by the tmr curial barometer. See BAROMETER_ To get the true pressure of the air it necessary to apply corrections for tempera: (32° F. is the standard), gravity, which vans. with latitude, and instrumental error; and whet barometer readings from several station; r' to be compared it is required to reduce den to what they would be if the stations all WI common level. Usually they are reduced tr sea-level by adding to each the inches of tar. cury equivalent in weight to the column of between the barometer cistern and sea-lera This is the pressure usually entered on weaite charts.