22 THE ATMOSPHERE. II. 



above the sea-level. At a height of 5520 metres (i.e., about 

 18,110ft.) the pressure is half what it is at the sea-level, while 

 at 11,040 metres (or 36,220ft.) it is reduced to one-fourth of 

 the sea-level pressure, and so on. In the case of small eleva- 

 tions it may be said that, roughly, an ascent of 900ft. lowers 

 the barometric pressure by an inch. From observations on 

 luminous meteorites and refraction of sunlight it has been esti- 

 mated that at a height of 200 miles air has a very small but 

 appreciable density. It is almost impossible to conceive of an 

 actual, denned limit to a gaseous atmosphere. 



The pressure of the atmosphere varies, as is well known, 

 from day to day, and upon the variations of pressure the direc- 

 tion and force of the w r ind depend. 



From a consideration of the mean pressures of many diffe- 

 rent parts of the world it has been discovered that there are 

 two broad belts of high pressure, one north and the other 

 south, running roughly parallel to the equator. Near the 

 equator itself, and also near the poles, are regions of low mean 

 pressure. 



Another fact in connection with atmospheric pressure has 

 been clearly noticed, most distinctly in tropical districts a 

 diurnal variation, there being two maxima, often about 9 a.m. 

 and 9 p.m., and two minima about 3 a.m. and 3 p.m. So 

 regular is this variation in some hot countries that, according 

 to Humboldt, it is almost possible to tell the time of day by 

 reading the barometer. 



The consideration of the relative heights of the barometer 

 obtaining at different places at a given time affords one of the 

 most important criteria in forecasting weather. 



Though this subject to the farmer is obviously of the greatest 

 importance, and though considerable progress has recently 

 been made in connection with it, this is not the place in which 

 it can suitably be discussed. The reader should consult a 

 modern treatise on meteorology. 



Dry air is almost diathermanous, i.e., transparent to heat 

 rays. Consequently it allows the sun's heat to pass through 

 with but little loss, becoming only very slightly warmed. It 

 also allows a considerable, though probably much less, per- 

 centage of the heat radiated from the earth to pass through it. 



