276 LECTURE XXII. 



shaken in it for a considerable time, the tube being held in an inverted posi- 

 tion; and where great accuracy is required, the mercury must be boiled in 

 the tube. The reservoir most commonly employed is a flat wooden boxy 

 with a bottom of leather ; the cover, which is unscrewed at pleasure, being 

 cemented to the tube. Sometimes a screw is made to act on the leather,, by 

 means of which the surface of the mercury is always brought to a certain level, 

 indicated by a float, whatever portion of it may be contained in the tube; 

 but the necessity of this adjustuicnt may be easily avoided, by allowing the 

 mercury to play freely between two horizontal surfaces of wood, of moderate 

 extent,aii I at the distance of one seventh of an inch: the height may then be 

 always measured from the upper surface, without sensible error. But if the 

 surfaces were closer than this, the mercury would stand too high in the tube. 

 (Plate XIX. Fig. 254.) 



The same method which is employed for determining the relation between 

 the heights and densities of elastic fluids, may be extended to all bodies 

 which are in any degree compressible, and of which the elasticity is subjected 

 to laws similar to those which are discoverable in the air and in other gases: 

 and it is not improbable that these laws are generally applicable to all bodies 

 in nature, as far as their texture will allow them to submit to the operation 

 of pressure, wi^'hout wholly losing their form. Water, for example, has 

 been observed by Canton to be compressed one twenty two thousandth of 

 its bulk by a force equal to that of the pressure of the atmosphere ; consequently 

 this force may be represented by that of a column of water 750 thousand feet 

 in height; the density of the water at the bottom of a lake, or of the sea, will 

 be increased by the pressure of the superincumbent fluid ; and sui)posing the 

 law of compression to resemble that of the air, it may be inferred that at the 

 depth of 100 miles, its detisity would be doubled; and that at 200 it would 

 be quadrupled. The same measures would also be applicable to the elasticity 

 of mercury. But there is reason to suppose that they are in both cases a 

 little too small. 



