EFFECT OF DRYING SOILS ON WATER-SOLUBLE CONSTITUENTS 201 



regards hygroscopic moisture and surface having, respectively, 1.25 and 0.90 

 per cent hygroscopic moisture. However, when exposed to a saturated atmos- 

 phere for 7 days, they stand in a different relationship to each other, their 

 positions being reversed, and instead of the soil with greater surface absorbing 

 more moisture, more is taken up by the soil which has less surface, but con- 

 tains more soluble salts. This tends to show that within certain limits, at 

 least, soluble salts are of sufficient importance to reverse the normal effect 

 of total surface exposed by a soil. 



Considering table 7 in a broad, general way, the hygroscopic moisture con- 

 tent in the air-dry soil bears a normal relationship to the total surface, and 

 when we consider surface and organic-matter content this same general re- 

 lationship holds for hygroscopic coefficient except in case of the sandy loam 

 where the discrepancy is due to its large quantity of soluble salts. The dif- 

 ficulty here lies in that we have not a single variable factor, but several, viz., 

 size of particles, total surface, organic matter, and soluble salts. 



Since it is almost, if not wholly impossible to control all factors at once, 

 it may be necessary to work with synthetic soils, varjdng but a single factor 

 at a time. These determinations of hygroscopic coefficients cannot yield much 

 valuable evidence, except that, in a general way, the soils with the highest 

 percentage of clay and colloids have the highest hygroscopic coefficients and 

 within the limits of this experiment have the highest soluble salt content. 

 However, the data are so meagre that no definite safe conclusion may be drawn. 



Experiment 6. Retention of nitrate by quartz and white silt loam 



The purpose of this experiment was to study the question whether clean 

 quartz sand holds potassium nitrate, and if so, to what extent. This salt 

 was selected because the acid radical is not generally supposed to be adsorbed 

 to an appreciable degree, and also because it was available as a chemically 

 pure substance. 



The material used was a clean ground quartz. It was first leached with a 

 10 per cent solution of hydrochloric acid to remove any possible soluble ma- 

 terial, then washed free of the acid and air-dried. It was then sifted into four 

 grades, as used by the United States Bureau of Soils: 



Diameter 

 mm. 



Coarse sand 1 .00-0.50 



Medium sand 0.50-0.25 



Fine sand 0.25-0.10 



Very fine sand 0. 10-0.05 



White silt loam from experiments 4 and 5 was used also, since in the air-dry 

 condition it was essentially devoid of nitrates. 



The water-holding capacity of 50 gm. of each of these materials was then 

 determined experimentally. Water was added from a burette and the ma- 

 terials were found to hold, without any loss by percolation on standing, the 



