THE MAINTENANCE OF SOIL FERTILITY 555 



of a soil is obviously impossible unless some knowledge is at 

 hand regarding the losses and additions which a soil sustains 

 in the course of a definite ^r otation. Fortunately, some fairly 

 reliable data have already been presented regarding the re- 

 moval of soil constituents under controlled conditions. The 

 Cornell lysimeter tanks, bearing a rotation of maize, oats, 

 wheat, and two years of hay, offer very satisfactory informa- 

 tion (paragraphs 95 and 163). The losses covering a ten-year 

 period are expressed in pounds to the acre a year. The soil is 

 a Dunkirk silty clay loam. 



While such figures are probably open to considerable error 

 and obviously would not apply with any degree of accuracy 

 to a light soil, they indicate in a general way the magnitude 

 and order of the losses that may be expected from such a soil 

 under the conditions specified. 



Table CXXVIII 



LOSSES FROM A DUNKIRK SILTY CLAY LOAM SOIL EXPRESSED IN 



POUNDS TO THE ACRE A YEAR OVER A TEN- YEAR PERIOD. 



ROTATION: MAIZE, OATS, WHEAT AND TWO YEARS 



HAY. CORNELL LYSIMETER TANKS. 



The organic carbon in this soil over the ten-year period was 

 reduced at the rate of approximately 1 per cent, a year. 2 This 

 is equivalent to a reduction in organic matter of about 1200 



x The largest loss of carbon is probably to the atmosphere as carbon 

 dioxide. The other avenue of loss is in the drainage water. 



a Lipman and Blair report a reduction of organic carbon of .74 per 

 cent, a year over a period of ten years on Sassafrass loam in New 



