2 THE AIR 



height of some 300 miles ; but, so far as meteorology 

 is concerned, 20 miles may be taken as the limit. 



The passage of air across a continent is a complex 

 of horizontal and vertical movements, in a medium of 

 continually varying density ; but over the far wider 

 stretches of ocean the density is nearly uniform, vary- 

 ing by but a small fraction of its maximum amount in 

 a horizontal direction, and falling off uniformly up- 

 wards. The oceans are, in fact, the only parts of the 

 Earth's surface where it is possible to study the 

 atmosphere in fairly normal conditions. The practical 

 uniformity of level eliminates differences of pressure 

 due to gravity ; the uniform composition of the water 

 surface eliminates the irregular heating and cooling 

 of the air. These last are due to the different specific 

 heat, conductivity, transparency, and power of radia- 

 tion and absorption, which characterize the rocks, sand, 

 soil, lake surfaces, and stretches of vegetation, which 

 diversify the land, causing them to react differently 

 and in various degree to the radiation which is received 

 from the sun, or is emitted from the Earth. The 

 homogeneity of the substance and form of the ocean 

 surface favours the normal development of atmospheric 

 equilibrium and adjustment, and gives to marine 

 meteorology a large simplicity, more favourable to the 

 elucidation of the great laws of atmospheric circulation 

 and disturbance, than is to be found in the more varied 

 meteorology of the land. Indeed, it is the effort to 

 constitute by computation some approach to the simple 

 conditions of the oceans which leads the meteorologist 

 to reduce to their equivalent values at sea-level, the 

 pressures and temperatures which have been observed 

 on the land, before the data are considered to be in a 

 state fit for charting in the ordinary maps of isobars 

 and isotherms. Being free from a multitude of dis- 



