Okfervations on Barometers. [Book V. 



under the form of elaftic fluids, diminish the weight of 

 the column of air, becaufe they are lighter than it ; but 

 thofe fubftances which are foluble in air augment its 

 denfny, and confequently its weight, in the fame man- 

 ner that fah diflblved in water increafes its weight and 

 den!lty. 



Tiie barometer has, therefore, another property, not 

 lefs ufeful to philofophers than that which has been 

 already mentioned. It points out the changes of the 

 weather, efpecially when they are likely to be confi- 

 derable. 



From the numerous obfervations and experiments, 

 which have been made from time to time upon baro- 

 meters, the feveri following propofitions have been efta- 

 bli/hed by M. BriiTon. " Firft, That the mean height 

 of mercury in France is twenty-feven French inches 

 and an half. Secondly, That the variations from this 

 height feldom exceed three inches, that is, that its leaft 

 elevation is twenty- fix inches, and its greateft twenty- 

 nine. Thirdly, That thefe variations become lefs 

 towards the equator, and greater in the northern cli- 

 mates. Fourthly, That when the mercury falls in the 

 barometer it announces rain or wind, or in general what 

 is called bad weather. Fifthly, On the contrary, when 

 the mercury rifes it announces fine weather. Sixthly, 

 That thefe predictions fail fometimes, efpecially if the 

 variations in the height of the mercury are very flow 

 and inconfiderable. Seventhly, That the predictions 

 are almoft infallible, when the mercury afcendsor de- 

 fcends confiderably in a Ihort time j as for example, 

 about one-third of an inch (or three or four lines) ia 

 the courfe of a few hours *." 



Thus in relating the difcovery of the barometer, we 

 have feen that philoiophers were convinced that an 



* Brifibu. Vol. i. 



aftual 



