The changes in the absorptive power brought about by wetting and drying, as shown in the above table, are very insignificant. In the case of the charcoal, soil and silt the diminution may fairly be assumed to be caused by the deposition of soluble salts on the surface, partly clogging the pores. In the case of the clay as well as in that of the river silt, the inevitable content of organic matter in process of decomposition has doubtless influenced the result, as is suggested by the increase of carbonic gas. That prepared chalk should in one case contain exclusively nitrogen gas, in the other case mixed gases, seems to indicate a difference in the air to which it is exposed, or in the water employed in its preparation; the latter case agreeing substantially with the results obtained from the precipitated carbonate. In both (as well as in the carbonates of barium and strontium), the absorption of carbonic gas is very small, or _nil_. It thus appears that for the condensation of carbonic dioxid gas, ferric and aluminic hydrates are prepotent among mineral substances; while clays, river silts and soils may always be expected to contain relatively large proportions of this gas in absorption. THE AIR OF SOILS. _The Empty Space in Soils._—In dry soils the empty space, usually amounting to from 35 to 50 per cent of its volume, is filled with air;[100] in moist or wet soils the space unoccupied by water is similarly filled. Hence when soils are in their best condition for the support of vegetation (chap. 11, p. 202), about one half of their interstices is filled with water, the other half with air. Actual measurements of the amount of air contained in well-cultivated garden soil have been shown by Boussingault and Levy to range between 10,000 and 12,000 cubic feet per acre, substantially agreeing, therefore, with the above statement. In uncultivated forest soil, on the contrary, they found only from somewhat less than 4000 to 6000 cubic feet of air per acre. Extended observations since carried out by Wollny, Ebermayer, and others have in general confirmed the earlier observations, while adding greatly to their significance in respect to their relations to plant growth, and to the process of humification and soil-formation. [100] The normal composition of atmospheric air is given on p. 16, chap. 2. As a matter of course, when water evaporates from the soil in drying, its place is taken by air so far as it is not filled by capillary water drawn from below.