in Fluids. 395 



When these particles, on being heated by the sun- 

 beams, began to move, they first arose up nearly per- 

 pendicularly ; but before they had risen to any con- 

 siderable height, they were carried away obliquely and 

 nearly in an horizontal direction by the lower current, 

 answering to the wind which in the atmosphere prevails 

 at the surface of the earth. 



The perpendicular rise of these particles from the bot- 

 tom of the box, and the subsequent change of their di- 

 rection, called to my remembrance an appearance very 

 common in hot countries, which I recollected to have often 

 seen, and by which I had often been amused in my 

 youth : in very hot and dry weather, when the wind is 

 still and the sun very powerful, the air which lies on the 

 ground often appears in the most violent agitation, re- 

 sembling that of a boiling liquid; which motion is most 

 rapid at the surface of the earth, and appears to cease at 

 the height of five or six feet above the ground. 



Is not this violent agitation occasioned by the conflict 

 which takes place between the hot and the comparatively 

 cold air moving verticaJfy, and in opposite directions, 

 very near the surface of the ground ? And are not the 

 winds which prevail above occasioned by the efforts of 

 whole strata of air to ascend or descend obliquely ? 



The currents I observed to prevail in my artificial at- 

 mosphere were never perfectly horizontal ; and if my 

 suspicions with respect to the cause of the winds are 

 well founded, neither can those winds be horizontal which 

 prevail in the superior regions of the atmosphere of the 

 earth, though they may be very nearly so. 



The greatest velocity of the currents in the saline 

 liquid in this experiment was nearly two inches in a 

 minute, but their motions were in general much slower. 



