THE 0A8B0US PORTIONS OF THE ATMOSPHERE. 



23 



burning, makes for itself an envelope, but to see this envelope 

 it is necessary to be outside of it, and looking at it in profile. 

 To those beneath, it appears at first to be only a slight mist, 

 which afterwards darkens as condensation increases, until at 

 last a black thunder cloud may be formed. But the conden- 

 sation of vapour may then possibly have diminished, and the 

 rushing in of cool air may have partially restored the equili- 

 brium of pressure, and the barometer may have begun to rise 

 though the storm may still be raging. We may conclude, 

 then, that rapid expansion of the column of air by heat of 

 condensation, up to a considerable height in the atmosphere, 

 is the cause of the great and sudden falls of the barometer that 

 sometimes occur ; and those falls are likely to be the greatest 

 where the heating begins in the lowest, and consequently the 

 densest parts of cold regions, and ascends the highest in those 

 regions. The low heavy air of such parts is greatly expanded 

 by the heat of condensing vapour, and being near the surface, 

 the diminution of pressure that follows, affects only a small 

 area of that surface, within which however the diminution 

 is great. But as we proceed to warmer climates, though the 

 quantity of vapour is there much greater, yet, as condensation 

 commences at a superior elevation, a higher and rarer portion 

 only of the atmospheric gases is expanded by it. Within the 

 tropics, the whole of the lower stratum of the atmosphere is kept 

 warm by condensation of vapour, and extraordinary cases of 

 condensation occur only at considerable elevations, — the dimi- 

 nution of pressure is therefore experienced over a wide extent, 

 but in each particular part of that extent, only in a small 

 degree. 



While in the arctic regions, the whole of each particular 

 disturbance is within the view of a single observer, to another, 

 within the tropics, it often appears to have an unlimited range, 

 and seems to fill the whole atmospheric space to a great height. 

 A consequence of this is, that, in any one place within the 

 tropics, the barometer falls but little, seldom more than half an 



