THE AIR OF THE SOIL. 221 



much of it as may be expressed by pressure, is not nearly 

 saturated with this gas. 



De Siiussure {Recherches Chimiques sur la VegetatioUy 

 p. 168) filled large vessels with soils rich in organic mat- 

 ters, poured on as much water as the earth could imbibe, 

 allowing the excess to drain off and the vessels to stand 

 five days. Then the soils were subjected to powerful 

 pressure, and the water thus extracted was examined for 

 carbonic acid. It contained but 2"|„ of its volume of the 

 gas. 



Since at a medium temperature (60° F.) water is capa- 

 ble of dissolving 100" |„ (its own balk) of carbonic acid, it 

 would appear on first thought inexplicable that the soil- 

 water should hold but 2 per cent. Henry and Dalton long 

 ago demonstrated that the relative proportion in which 

 the ingredients of a gaseous mixture are absorbed by wa- 

 ter depends not only on the relative solubility of each gas 

 by itself, but also on the i)roportions in which they exist 

 in the mixture. The large quantities of oxygen, and 

 especially of nitrogen, associated with carbonic acid in the 

 pores of the soil, thus act to prevent the last-named gas 

 being taken up in greater amount ; for, while carbonic 

 acid is about fifty times more soluble than the atmos- 

 pheric mixture of oxygen and nitrogen, the latter is pres- 

 ent in fifty times (more or less) the quantity of the former. 



Absorption of Carbonic Acid by the Soil. — ^According to 

 Van den Broek, {An>u derChemie n. Ph., 115, p. 87) certain 

 wells in the vicinity of Utrecht, Holland, which are exca- 

 vated only a few feet deep in the soil of gardens, contain 

 water which is destitute of carbonic acid (gives no precipi- 

 tate with lime-water), while those which penetrate into the 

 Underlying sand contain large quantities of carbonate of 

 lime in solution in carbonic acid. 



Van den Broek ina<le tlie Ibllowing experiments with 

 garden-soil newly manured, and containing free carbonic 

 acid in its interstices, -wiiich could be displaced by a cur* 



