March 7, 1890.] 



SCIENCE. 



163 



■we will now examine how this changes for different tempera- 

 tures and at different levels. The following table is gathered 

 from Glaisher's "Hygrometrical Tables" (IV. and VI.) : — 



Weights in Grains of One Cubic Foot, at 30 Inches Pressure. 



It will be seen from this table how much lighter vapor is than 

 air, and that the difference in specific gravity is highly increased 

 as the temperature sinks. While air follows Gay Lussac's law 

 by expanding by heat and contracting by cold, the vapor follows 

 •a law the reverse of Gay Lussac's by contracting by heat and ex- 

 .panding by cold, but at a much greater rate for equal tempera- 

 tures: ~ gives us the specific gravity of vapor, tliat of the surround- 

 ing air being 1 ; it shows how the buoyancy of vapor I -- 1 is 



.strongly increased as the temperature sinks. At 100° the specific 

 gravity of vapor is one-twentieth of that of the surrounding air, 

 and at 0° it is only a thousandth part of it. The weights are 

 measured under a pressure of 30 inches, the pressure at the earth's 

 surface. To find the weights at higher levels, where the pressure 

 is less, we have only to multiply the numbers of columns (a) 

 and (u) with the same factor according to Marriott's law. As 



- hereby remains unchanged, it appears that the buoyancy of 



vapor in the atmosphere depends entirely upon the existing tem- 

 peratures, and is independent of the pressure, or the level at 

 which the vapors are found. As the temperature constantly 

 sinks as we rise in the atmosphere, the buoyancy of vapor, or 

 the force with which vapors tend upwards as they rise to higher 

 levels, is constantly increased, and at an astonishingly high rate. 

 "While, therefore, the speed with which vapors rise in the at- 

 mosphere may be more or less imperceptible at the ordinary tem- 

 peratures at the earth's surface, it is rapidly increased as the 

 vapors rise, and may attain an almost inconceivable magnitude 

 in the extreme cold which exists at a great distance from the 

 earth's surface. 



With the results deduced from the table fresh in our mind, 

 ■we may now draw a picture from nature while trying to follow 

 the vapors on their upward passage tlirough the atmosphere, and 

 ■we shall see how far our calculations agree with the natural 

 phenomena. To take a distinct case before us, let us suppose 

 that on a fine day, with high barometer, we are in a dry locality 

 in which is found an isolated swampy place or lake (Fig. 2) . 

 While the surface-air is dry generally, we find it moister over 

 the swampy place, as the sun and the warm and dry air which 

 passes over it cause a strong evaporation to take place. The 

 w^arm surface-air, though expanded by heat, moves over the 

 ground without rising. It is first caused to ascend by being in- 

 termixed with the vapor-particles. According to their buoy- 

 ancy, the vapor-particles tend upwards through the atmosphere, 

 thereby carrying the air with which they are intermixed up- 

 wards also, and the ascent of a current of damp air is established. 

 The vapors are the real cause or life in this motion, each and all 



of its particles acting as so many minute balloons. Some eight 

 or ten thousand feet overhead, perhaps at a little distance later- 

 rally from the moist ground, according to the direction in which 

 the air moves over the ground, we notice an enormous cumulus- 

 cloud being formed, and we have no doubt whatever that it is 

 caused by the current of damp air ascending from the moist piece 

 of ground. Tlie ascending cun^ent, after having passed through 

 the heated surface-air, gets suddenly into a much colder stratum, 

 and condensation takes place by mixture of the rising damp air 

 with the cold air it is passing through. As a rule, the chilling 

 caused by the expansion of the ascending current gives it a tem- 

 perature pretty nearly the same as that of the air through which 

 it passes. It is only when it is met by a sudden change in the 

 temperature of the surrounding air that condensation takes place 

 by mixture, which we may express by saying that the ascending 

 current has ' 'caught a cold. ' ' Instead, therefore, of it being the 

 chilling by expansion which causes condensation into clouds and 

 thereby rain, we see that it is a fact that the chilling was not 

 sufficient when the ascending current was taken by surprise by 

 the sudden change in temperature If the colder sti^atum of air 

 be moving along, we may notice a row of detached cumulus- 

 clouds at some distance from the one nearest to the moist piece 



of ground, but they grow smaller and smaller the farther away 

 they pass. They are thus cut off from the supply of damp air, 

 and being surrounded by unsaturated air on all sides, and ex- 

 posed to the sun's rays, they rapidly evaporate. 



The formation of these cumulus-clouds was therefore only a 

 passing event in the ascent of the current of damp air ; and as the 

 vapors rose before they were condensed, so they will rise again 

 when they are turned into invisible vapor again, and the more 

 quickly, the faster the temperature sinks during the ascent. 

 While, therefore, air and vapor are equally expanded by decrease 

 of pressure during ascent, the decrease of temperature acts dif- 

 ferently upon them, having the effect of contracting the air, 

 while the vapors are very much expanded. For both these 

 reasons the buoyancy of the vapor is increased during the 

 ascent. Tlie vapors must therefore necessarily rise as long as 

 there is any air to pass through, unless they meet with a layer of 

 saturated moisture, or air saturated with moisture. 



The clouds produced by the ascent of a current of damp air are 

 cumulus-clouds, and they resemble in their shape very much the 

 mist caused by steam escaping from a chimney. The phenomena 

 are, in fact, precisely similar; and the cumulus-clouds are in 

 their nature as unstable a product as the mist from a chimney , 

 only the first phenomenon is on a much larger scale, and con- 

 sequently it takes a much longer time for the cumulus-clouds to 

 evaporate. 



