386 Journal of Agricultural Research vol. xx. no. s 



a soil having a low percentage of colloidal material and having a fairly __ 

 high concentration in its soil solution, which exercises a repressive effect 

 on the solubility of certain soil constituents as previously explained. 



In Table I the results of the equilibrium studies with three different 

 soils are presented. Comparisons are made between the composition of 

 the concentrated extracts and the same extracts after treatment with 

 the soil. It will be noted that the total concentration has suffered prac- 

 tically no change, as shown by the freezing-point depressions, conductivity 

 determinations, and proportion of total solids. Also, the concen- 

 trations of potassium, magnesium, calcium, nitrate, and sulphate agree 

 within the limits of experimental error. The agreement for sodium is 

 less perfect, but considering the small quantities involved the differences 

 are also probably within the limits of error. In one case more phosphate 

 is found in the re-extract, and in two cases the agreement is fairly close. 

 In one case the two silica determinations agree almost perfectly, and in 

 two cases silica seems to have been retained by the soil. It is very 

 difficult to explain the action of this radicle, first because of the chance of 

 contamination of the solution from glass vessels and secondly because of 

 the numerous types of silicates possible with varying proportions of 

 silica. 



While the agreement between. the extracts and re-extracts is on the 

 whole remarkably close, it might be objected that the conditions for the 

 attainment of equilbrium were inadequate and that another extract 

 having a different composition might also remain unchanged by the soil. 

 In order to test this possibility extracts were made of soils 9 and 15 in 

 the previously described manner, and then potassium sulphate was added 

 to the extracts so as to double approximately the concentration of 

 potassium. These modified extracts were then concentrated until they 

 had the same osmotic value as the soil solutions, and re-extracts were 

 made as in the first experiment. The composition of the different 

 solutions is given in Table II. It is evident that in this experiment the 

 soil has had a marked effect on the extract. There is very much less 

 potassium in the re-extract than in the original extract, but the decrease 

 of potassium is accompanied by an increase in the quantity of calcium 

 and in one case of sodium. In one case there is a slight decrease of 

 sulphate. The other elements are not greatly changed, nor is the total 

 concentration very different in the two cases. It seems clear that a 

 rearrangement of the solutes has taken place in this case which did not 

 occur in the first experiment. In other words, the extract introduced 

 was different in composition from the soil solution already present, with 

 the result that certain chemical reactions took place forming an entirely 

 new soil mass — soil solution svstem. 



