NATURAL HISTORY. 47 



sume more of a winding course. In large rivers there 

 is a considerable eddv along the banks ; and the near- 

 er the sea this eddy is the greater. The surface of 

 the water in rivers is by no means level from bank to 

 bank ; on the contraiy, the middle of the stream is 

 higher or lower than the water of the sides according 

 to circumstances. When a river swells suddenly by 

 the melting of snow or any other cause, the middle of 

 the stream is sensibly higher than the sides : in one 

 instance the elevation is said to have been three feet. 

 On the other hand, when rivers approach their mouths 

 the water near the sides is commonly more elevated 

 than in the middle. 



The inundation of the Nile, though nothing can be 

 more natural, has long afforded matter for curious 

 and doubtful speculation. It is the rain which falls 

 in Abyssinia and Ethiopia which occasions the swel- 

 ling and inundation of this river, though the north wind 

 must be regarded as the primitive cause. 1st, Be- 

 cause it drives the clouds which convey this rain from 

 the coast of Abyssinia: 2dly, Because blowing a- 

 gainst the two mouths of the Nile, it forces the wa- 

 ters back against the stream, and thus prevents them 

 from pouring into the sea in too great a quantity : 

 this circumstance may every year be relied on, when 

 the wind being at the north, and suddenly veering to 

 the south, the Nile in one day loses what it gathered 

 in four. 



Inundations are generally greatest in the superior 

 parts of rivers ; because the velocity of a river uni- 

 formly increases until it empties itself in the ocean. 

 But as the theory of running waters is subject toinany 

 difficulties, we ought carefully to study the peculiari- 

 ties of particular rivers. 



