Feb. II, i9i8 Freezing-Point Method and Soil Solution 391 



freezing-point depressions as a basis, it appears that from 0.0 1 to 0.03 

 gm. of total solids is in solution for each 100 gm. of moist soil, while 

 the total solids obtained by i-to-5 extractions vary from 0.02 to 0.06 

 gm. per 100 gm. of soil. 



In order to obtain approximate estimates of the material dissolved in 

 an extract additional to that present in the soil solution, two soils (No. 

 5 and 8) were subjected to the procedure of successive extraction. One 

 kgm. of each soil was extracted with a total of 10 liters of distilled water. 

 Ten extractions were made, with i liter of water for each extraction. 

 The results of the 10 extracts are plotted in figure 9. It will be noted 

 that the first extract contained a very much greater quantity of dissolved 

 solids than any subsequent extract. It was suggested by Stewart that 

 some idea of the extra dissolved material might be formed by using as a 

 basis of calculation the quantity dissolved in each liter after the extracts 

 had become relatively constant. Thus, for soil 5 roughly 150 p. p. m. of 

 total solids were found in each of the later extracts, and for soil 8, 100 

 p. p. m. By assuming that the rate of solubility is uniform, in a i-to-5 

 extract there would be 5 times 150, or 750, p. p. m. of extra material dis- 

 solved in the case of soil 5, and 500 p. p. m. for soil 8. The total solids 

 dissolved from i kgm. of soil by 5 liters of water aggregate 1,300 and 

 1,100 p. p. m. of total solids, respectively. This means that not more 

 than 50 per cent could have originally been present in the soil solution. 

 If this correction were applied to the extracts, the agreement with the 

 data obtained by the freezing-point method would in some cases be 

 fairly close, but usually the extracts would still give somewhat higher 

 results. This fact is reasonably explained on the assumption that the 

 first liter of solvent has dissolved out some material relatively soluble 

 although not actually present in the soil solution, while the later extracts 

 contain only difficultly soluble substances. 



GENERAL DISCUSSION 



It will now be well to correlate certain broad relations which may be 

 deduced from the data presented in this paper through some further 

 reference to the principal investigation of this series by Stewart. It 

 should be emphasized again that neither from the water extraction nor 

 freezing-point methods is there any evidence that the soil solution has a 

 constant composition. On the contrary, the soil solution even of the 

 same soil may vary greatly under different conditions. The water- 

 extraction method indicates that this variation may occur with all the 

 principal nutrient elements with the exception of phosphorus. These 

 data accord with the view expressed by Bouyoucos and McCool, that 

 the soil solution should not be considered as saturated. The opposite 

 conception as outlined in the earlier work of Whitney and Cameron (9) 

 is not upheld by the present investigation. If the great excess of nearly 

 insoluble minerals in a soil were the determining factor in the soil solu- 

 27809«— 18 7 



