by masses of extremely cold air that constantly are thrown upward against it.
The humidity of the air of the stratosphere is very low—so low that visible clouds do not form. Therefore, if the dew-point is ever reached, the condensation is confined to ice spicules so few in number that they do not affect the visibility of the air. There is no vertical convection; therefore they sink slowly; and if they are greater in size than are molecules of water vapor they sink more rapidly than the water vapor diffuses itself.
It seems certain that the air of the stratosphere contains dust a-plenty—both cosmic dust and dust that is hurled into it by volcanic eruptions. If dust is absent, the air of the stratosphere differs from that below it and from space above it. One thing is certain, the radio-activity within the stratosphere indicates the presence of dust particles highly electrified.
The depth of the troposphere is inconsiderable compared with that of the stratosphere; aviation has probably scaled its height probably within pistol shot distance of the isothermal layer. The troposphere is the region of convection. Its height is practically the height of cirrus clouds, and all the great movements of the air—wind, cloud, storm, and precipitation—take place within its limits.
Experience has taught the meteorologist that conditions in mid-air of the troposphere, in many instances, are the key to conditions at the surface. They are far more important in air flight; for the airman encounters bumps and holes, both of which are due to sudden inequalities in temperature. The airman and the navigating officer of the airship are likely to encounter cross-winds, the updraught of thunder-storms, and the vagaries of cloud-formation; these, too, are due to irregular conditions of temperature, all of which must be understood and reckoned with in flight.
Air Altitudes and Terrain Altitudes.—The laws and values which apply to vertical altitudes in free air are not applicable to altitudes on the earth’s surface. In general, temperature decreases with altitude, but this is not always true. At various times the temperature of mountain valley floors is lower than that of the foot-hill slopes several hundred feet higher. On very cold, still nights, low spots, such as stream valleys, are almost always colder than higher ground. In regions where