[ 38o ] 



Again, another difference muft be remarked betwixt the di- 

 reflion of the trade winds below and that of the fuperior current. 

 The trade winds are chiefly eafterly, with only a few points to 

 the north or fouth, according to their fituation on the north or 

 fouth fide of the equator, commonly one or two, rarely more. 

 But the diredion of the upper current is chiefly to the north or 

 fouth, according to the tropic the fun approaches, with only a, 

 few points weftwards, as the greatefl cold prevails in the northern 

 or fouthern quarters. 



The height above the level of the fea or furface of the earth, 

 at which this intumefcence begins to overflow, is that at which 

 its denfity notably furpaffes that of the aggregate of the adjacent 

 extratropical columns at the fame height, that is about -j'j- or -jVth, 

 and even l^ill lefs. 



To ftate this point more clearly we muft take a general furvey 

 of the temperatures of the different aggregates of air thus com- 

 pared, confining ourfelves to the northern hemifphere, as beft 

 known, and to the winter feafon. 



The 



