Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 339 



of solution, but the total solids were slightly increased as organic matter 

 replaced the nonvolatile solids. 



With Tejunga fine sandy loam 10 there was a distinct depression 

 throughout. Both solids and all individual constituents were decreased 

 in quantity. 



This evidence, together with that obtained by Hoagland (27) using 

 the freezing-point method, clearly showed that the differences which 

 may be found between cropped and fallow soils by the extraction pro- 

 cedure was always less than that which actually exists. 



In the same paper (27) Hoagland has also shown, both by special ex- 

 traction studies and by calculation from the amounts of total solids 

 contained in the i-to-5 extract, that the relation of the actual soil solu- 

 tion to this water extract is extremely close. The amounts of material 

 in the i-to-5 extract are approximately two to five times as great as in 

 the soil solution. The range of concentration shown by these extrac- 

 tions is therefore undoubtedly too high. This does not nullify the 

 differences between cropped and uncropped soils which are shown to 

 exist by this procedure. Hoagland 's results show that, if anything, 

 these differences are actually much greater. A decrease in the range of 

 concentrations would therefore have no more effect than to change the 

 scale upon which the seasonal curves have been drawn. 



EFFECT OF VARYING THE PROPORTION OF SOIL AND WATER 



It was important to learn what relation the conventional i-to-5 extract 

 bears to other possible ratios of soil and water. Reference has been 

 made to the work of Mitscherlich (42), in which he employed a series of 

 varied extractions of soil with water saturated with carbon dioxid. 

 This idea was now applied. Four soils, two silty clay loams and two 

 fine sandy loams, were extracted with proportions of soil to water vary- 

 ing from I part of soil to % part of distilled water up to i part of soil to 

 80 parts of distilled water. 



The results in parts per million of dry soil have been plotted in figures 

 3 to 6. It will be observed that with each soil the proportions of com- 

 pounds extracted in the lower ranges of concentration were essentially 

 the same. It was not until a range of concentrations of soil to water 

 greater than i to 10 were reached that a distinct change in the extrac- 

 tion of nutrients took place. These results therefore warranted the 

 employment of the i-to-5 extract as a conventional procedure. 



SEASONAL STUDIES OF THE WATER EXTRACT 

 FIRST SEASON, 1915 



In the first season's work, in 1915, both containers of each soil were 

 sown to barley. Half of the soils were planted on June 14, and the 

 remainder one week later. At the same time the first soil samples were 



