352 BAROMETER. 



warm southerly winds, which have been ex- 

 panded by heat and saturated with vapour, cause 

 the barometer to fall. It also follows, that as 

 rains are supplied chiefly by warm southerly 

 winds, in the middle and higher latitudes of the 

 northern hemisphere, the sinking of the baro- 

 meter must be an indication of falling weather : 

 and as it has been shewn, that all violent winds, 

 such as hurricanes, tornados, &c. are owing to 

 the rapid condensation of aqueous vapour, the 

 falling of the barometer must also forebode tem- 

 pestuous weather. It is equally evident, that cur- 

 rents of cold air raise the barometer, not only 

 because of their greater specific gravity, but 

 because they condense the steam of warm air. 

 Hence it is, that for two or three days before 

 storms of thunder and lightning, wind and rain, 

 during the prevalence of warm southerly winds, 

 the barometer falls ; but rises during cool and 

 serene dry weather.* 



* From several passages in Mr. Daniell's work, on the atmos- 

 phere, it would seem that he referred the sinking of the baro- 

 meter to the evolution of caloric from aqueous vapour during its 

 condensation, by which the atmosphere is expanded where it 

 takes place. (See page 126, &c.) But it has been shown, that 

 within the tropics, where the amount of condensation is greatest, 

 the fluctuation of the barometer is very small ; while it is obvious 

 from the foregoing facts and observations, that the barometer is 

 depressed by the accumulation of aqueous vapour in the atmos- 

 phere, and elevated by its condensation. 



Dr. Dalton ascertained by a series of observations continued 

 for five years at Kendal, in the north of England, that the 

 barometer fell below twenty-nine inches during forty days ; only 



