Hoagland — 158 — Plant Nutrition 



increasing concentrations of potassium in the soil 

 solution, even though most of the added potassium 

 is fixed by the soil colloids. As the potassium incre- 

 ments are made larger, the percentage fixation of 

 potassium diminishes, as would be expected. Whether 

 the plant utilizes replaceable potassium directly (con- 

 tact effect) or through the medium of the soil solu- 

 tion, cropping tends to reduce the concentration of 

 potassium in the latter. While, as already explained, 

 plants absorb more potassium proportionately to con- 

 centration from solutions of decreasing concentration, 

 within certain limits the absolute amounts absorbed 

 per unit of plant tissue increase with increasing soil 

 solution concentrations (and increasing values for 

 replaceable potassium) and so the percentage content 

 of potassium in the plant rises to a maximum value. 

 This value depends on physiological and genetic 

 factors. 



These explanations do not cover all the pertinent 

 phenomena. There are soils that supply enough potas- 

 sium for the production of excellent crops, which 

 nevertheless contain only very small amounts of re- 

 placeable potassium at any time, amounts which do 

 not decrease with continued cropping; and at the 

 same time the soil solution concentrations are low. 

 We then say that the crop is, in effect, deriving 

 potassium from non-replaceable, or very difficultly 

 replaceable, form. The evidence for this statement 

 is extensive and includes results obtained by this 

 laboratory in a greenhouse experiment in which nu- 

 merous soils were successfully cropped over a period 

 of years, with quantitative determinations of the potas- 

 sium withdrawn from the soils by the plants and 

 chemical studies on the soils themselves. 



The conclusions have a practical bearing on at- 

 tempts to ascertain the "available" potassium in soils 

 by simple chemical examinations, as by using dilute 

 acids to extract potassium. The point will perhaps 

 be made clearer by citing an experiment by Martin 



