INTRODUCTION. 4^ 



!?*rom thefe obfervations it appears that it is not fa 

 .much the height of the mercury in the tube, that indi- 

 cates the weather, as the motion of it up and dov.n. 

 Wherefore, to pafs a right judgment of \vha<- v/eaiher 

 is to be expeded, we ought to know whether ttv^ 

 mercury is exactly rifmg or falling ; for which end 

 the following rules are to be obferved : 



If the furface of the mercury is convex, (landing 

 .higher in the middle than at the fides, it is generally 

 a lign that the mercury is rifmg. 



If the furface of the mercury is concave, or hollow 

 in the middle, it is fmking. 



If it is plain or level, or, rather, if it is a little con- 

 vex, the mercury is flationary; for mercury being 

 put into a glafs tube, efpecially a fmall one, will na- 

 turally have its furface a little convex; becaufe the 

 particles of mercury attract each other more forcibly 

 ,than they are attracted by glafs. 



If the glafs be fmall, fhake the tube, and if the air* 

 be grown heavier, the mercury will rife about half 

 the tenth of an inch higher than it flood before ; if 

 it is grown lighter, it will fmk fo much. This pro- 

 ceeds from the mercury flicking to the fides of the 

 tube, w^hich prevents the free motion of it, until it is 

 difengaged by the fhock. Therefore, when an obferv- 

 ation is to be made by fuch a tube, it ought ahvays 

 to be fhaken firfl ; for fometimes the mercury will 

 not vary of its own accord until the weather it ought 

 to have indicated be prefent. Dr. Derham fays that 

 the variation^ of the barometer are greatefl neareil 

 the pole J in places near the equinodial there is 



fcarce 



