THE CHEMISTRY OF THE ATMOSPHERE. 95 



learned in Part L, Chapter VI., that the atmosphere is held 

 as a robe around the earth by the attraction of gravitation. 

 Now it is with the gases as with solid and liquid substan- 

 ces each is attracted to the earth in the proportion of its 

 specific gravity, the heaviest always taking the lowest posi- 

 tion that is, getting the nearest to the earth. As mercury, 

 therefore, gets below water, and water below oil, so car- 

 bonic anhydride tends to get below oxygen, and oxygen 

 below nitrogen. See what would be the consequence if 

 this tendency were allowed to be carried out unopposed. 

 The carbonic anhydride would be accumulated beneath all 

 the oxygen and nitrogen, filling up all the valleys, and lying 

 along upon all the plains. And as this gas is a deadly poi- 

 son, no animal could live except upon elevated places, hills, 

 and mountains. But even there life would be short, and 

 attended with suffering ; for the nitrogen, being lighter than 

 oxygen, would be above it, so that animals would breathe 

 air that would be too stimulating, producing fevers and in- 

 flammations, and the extreme readiness with which every 

 thing would burn would occasion constant trouble. 



120. Disposition of Gases to Mingle Together. The in- 

 fluence of gravitation upon the gases is counteracted to a 

 great extent by a disposition which we find in gases to 

 mingle with each other, and thus the disastrous consequences 

 above alluded to are prevented. The following experiment 

 beautifully exhibits this diffusion, as it is termed : Let a bot- 

 tle (Fig. 29, p. 96) be filled with carbonic anhydride, having 

 a long tube fitted into the cork. At the upper end of the 

 tube place a bottle of hydrogen gas. As the carbonic an- 

 hydride is twenty-two times heavier than hydrogen, gravi- 

 tation tends strongly to keep the carbonic anhydride in the 

 lower bottle, the hydrogen of course remaining in the upper 

 one. But observe what happens. If the apparatus be left 

 to stand for an hour or two, it will be found that there is a 



