FIELD EXPERIMENTS IN THE SOUTH 479 



A study of Table 23 will show that a 6 1 -bushel crop of corn re- 

 quires more potassium than a 3o-bushel crop of wheat, which fact 

 may account in part for the greater effect of potassium on corn, 

 although about the same relation holds for phosphorus. A more 

 important difference probably exists in the relative feeding powers 

 of the two crops, influenced (i) by the difference in root system, 

 including the different depths of feeding, (2) by the difference in 

 seasonal conditions and consequent difference in decay of humus, 

 in decomposition of other soil materials, and in activity of soil 

 organisms during the principal period of growth, (3) by the sol- 

 vent action of the carbon dioxid excreted by the bacteria and from 

 the plant roots, and (4) possibly by different requirements as to 

 the forms or combinations in which the plant-food elements can 

 be absorbed and assimilated or utilized by corn and wheat. 



The Rothamsted data contribute much toward the solution of 

 this practical problem, but the very important question recurs, 

 whether more or less of the effect attributed to potassium may not 

 be due to the stimulating action of the soluble potassium salt in 

 liberating other substances from the soil instead of serving directly 

 as plant food; and, if so, would it be advisable and more profitable 

 to substitute some other less expensive material, such as kainit, 

 for the concentrated potassium sulfate used in these experiments ? 



It can also be stated that as an average of 56 tests (including 

 the use of twenty-five different varieties of corn) conducted in 

 1907 and 1908 on the Illinois experiment field near Fairfield in 

 Wayne County, on the same type of soil, an application of 200 

 pounds per acre of potassium sulfate, containing 85 pounds of the 

 element potassium and costing $5, increased the yield of corn by 

 5.4 bushels per acre; while 600 pounds of kainit containing only 

 60 pounds of potassium and costing $4, gave 9.9 bushels' increase. 

 These applications are made but once for a four-year rotation. 

 The kainit with 25 pounds less potassium produced 4.5 bushels 

 more corn than the sulfate. At 40 cents a bushel for corn, the kainit 

 has paid for itself. Kainit contains about 25 p'er cent of potassium 

 sulfate together with some 16 per cent of magnesium sulfate, 12 

 per cent of magnesium chlorid, and 33 per cent of sodium chlorid, 

 all of which are soluble salts; and the results plainly indicate that 

 the effects produced are due not solely to the element potassium, 



