GRAVITY AND LEVITY OF GASES. 249 



exactly equal to the loss of resistance, which 

 the medium of air in the receiver has sustained, 

 in consequence of the rarefaction which it has 

 undergone. By artificial means such as these, 

 the precise degree of difference which exists, 

 between the density and rarity of different 

 gases, may be ascertained ; the number of me- 

 tallic grains employed, are standards t)f mea- 

 sure, which are necessary to restore the coun- 

 terpoise, in the opposite scale, and constitute 

 the measure of the difference which exists be- 

 tween the density of the air in the bottle, and 

 the rarity of the air external to it ; it gives 

 the relative, but not the absolute weight. If 

 a perfect exhaustion could be effected, we 

 should, in that case, have the absolute, not the 

 relative weight: but as this event cannot be 

 accomplished, we have, in every case, the rela- 

 tive weight, and not the absolute one. 



Various experiments have been made to as- 

 certain the relative weight of different gases, 

 not only with respect to each other, but with 

 respect to the utmost state of rarefaction that 

 can be attained : it has been found, that a 

 pint of atmospheric air, inclosed in a bottle, re- 

 quires the measure of about? metallic grains, to 

 restore the equilibrium, that hydrogen gas is 

 the most rare, because the lightest and car- 

 bonic acid gas the most dense, because the 



