[ 411 J 



the evaporation) difperfe the clouds, and proceeding from coun- 

 tries then fomewhat heated, allow the fun's rays their natural 

 calefaflive effed. But why winds from oppofite points fliould 

 prevail in different years, cannot be known, until the contempo- 

 raneous ftates of the atmofphere between the northern tropjc 

 and the equator, are known. It is pofllble that frequent hurri- 

 canes and tornadoes, during which a quantity of air may be 

 deftroyed and converted into water, may demand an annual 

 fupply from the north^. and thus occafion our north and eatt 

 winds ; and the abfence of thefe phasnomena may occafion an 

 influx from the fouth^ if the north and eaft are fummoiied to a 

 different quarter, by fijmilar caufes, 



labie 



