84 Experimental Philosophy. [Lecture 7. 



darkness the moment of its setting. The length 

 of the twilight, therefore, at a given time, is in 

 proportion to the height of the atmosphere : or let 

 us invert this, and say, that the height of the 

 atmosphere is in some proportion to the length of 

 the twilight. This consideration led to an inves- 

 tigation (to which we shall recur when we treat 

 of astronomy) from which it has been inferred 

 that at the height of 45 miles, the atmosphere has 

 sufficient density to bend the rays of light. At 

 greater altitudes, the density is not sufficient to 

 occasion any perceptible effects. 



The density of the air, however, depends not 

 merely on the pressure it sustains, but on other 

 circumstances ; so that it varies even at the same 

 height in different parts, and in the same place at 

 different times, as is seen by the mercury in the 

 barometer rising to different heights, according 

 to the state of the weather. Heat in particular 

 was mentioned as a very powerful cause in rarefy- 

 ing the air. From this circumstance arises one of 

 the most striking and formidable of the atmo- 

 spherical phsenomena the WIND. Wind is no- 

 thing but a strong current or stream of air. 

 Whenever the air is heated by the sun, or by any 

 other means, it will be rarefied, and less able to 

 resist the pressure of the adjacent air, which will 

 consequently rush in "to restore the equilibrium," 

 to speak in the technical language of philosophy, 

 or, in plain terms, to reduce the rarefied part to an 

 uniform density with the other. This current of 

 air is sensibly felt near the door of a glass-house, 



