36 PHILOSOPHY OF STORMS. 



the time of pumping air into the nephelescope and of letting 

 it out, and to my astonishment I found, the rise of the mer- 

 cury after the discharge, constantly greater, as the time was 

 longer up to about twelve or fifteen days ; but beyond 

 that time the effect did not seem to be increased. It follows 

 from these experiments, that when air, saturated with va- 

 por, is confined in a glass vessel, air tight, and containing a 

 small portion of water, it will cease to be saturated to the 

 amount of about 4 or 5 degrees in fifteen days. 



Whatever may be the cause of this remarkable fact, so 

 contrary to all our notions since the experiments of Dalton 

 on the subject of the dew point, the following table of ex- 

 periments proves beyond all doubt that it is a fact. 



Does water or glass so attract the particles of aqueous 

 vapor in contact with them, as to condense some of these 

 particles on them, and bring down the dew point 4 or 5 

 degrees below the temperature of the water, and the air in- 

 cluded in the vessel? The fact that air near the surface of 

 the ocean has generally a dew point several degrees below 

 the temperature of the water, would seem to lead to an 

 affirmative answer to this question, more especially as it is 

 now known, that aqueous vapor does not permeate the pores 

 of atmospheric air, and cannot rise into the upper regions of 

 the air, in any way except by the motion of the air itself. 



