Dec. i, 1920 Relation of the Soil Solution to the Soil Extract 385 



same concentration and composition as the soil solution already present, 

 then it may be assumed that the resultant extract would have the same 

 composition and concentration as the original solution. On the other 

 hand, if the solution used were of different concentration or composition 

 a readjustment of the equilibrium should take place so as to produce a 

 different extract. It was decided to attempt an experiment based on 

 this hypothesis. 



Obviously, the preparation of a solution having the same composition 

 and concentration as the soil solution is a matter of great difficulty. 

 The only feasible scheme seemed to be the use of a soil extract con- 

 centrated to a point where it would have the same concentration as the 

 soil solution, this concentration being determined by the method of 

 Bouyoucos and McCool. It was reasonable to assume that in such a 

 solution there would exist, between some of the most important elements, 

 a relation very similar to that found in the actual soil solution— that is, 

 the solution of the free water with the soil at approximately optimum 

 moisture content. In order to limit as far as possible the quantity of 

 solutes dissolved from the soil mass, an extract was made with cold 

 water, and only X part of water was used to 1 part of soil. The time of 

 contact was limited to that necessary for complete admixture. Filtra- 

 tion was made through a Buchner funnel, and final clarification was 

 effected with the use of a Pasteur filter. A separate portion of the soil 

 was then made up to its optimum water content, and the freezing-point 

 depression was carefully determined. The extract of the soil made in 

 the manner described was then concentrated on a hot plate, meanwhile 

 passing through the solution a stream of carbon-dioxid gas in order to 

 prevent any precipitation. Finally the volume of the concentrated ex- 

 tract was adjusted with distilled water so that it had exactly the same 

 freezing-point depression as that of the moist soil. This solution was 

 used in extracting the moist soil (1 part of soil to }4 part solution). Care- 

 ful analyses were made of the extract before and after contact with the 

 soil, and the results were compared. 



Before the data are considered it should be recalled that ordinarily in 

 a water extraction from 2 to 5 times as much total solids are dissolved 

 as are actually present in the soil solution, and this is true with the 

 extractions now considered. Under certain conditions, however, it is 

 possible to obtain an extract which contains a comparatively small 

 quantity of dissolved substances in addition to that originally present 

 in the soil solution, as indicated by the method of Bouyoucos and McCool. 

 For example, a sample of soil 9, having a freezing-point depression of 

 0.148 C. at 17 per cent moisture gave in a 1 to yi extract only about 1.16 

 times the quantity of total dissolved solids equivalent to this depression. 

 In this case the unfree water was determined directly by dilatometer 

 measurements (1). Such a result apparently can be obtained only with 



