THEORY OF SOIL FERTILITY 199 



not many records in the literature of yields 

 per acre for any field, farm, or country going 

 back for any considerable number of years. 

 Official estimates of yields have been kept by 

 the United States Government and many of 

 the States for the past forty years. Similar 

 statistical estimates have been kept by most of 

 the European countries for the past twenty or 

 thirty years. Beyond this the information is 

 fragmentary and possibly less trustworthy. 



"In the consideration of data of this kind 

 it would at first sight seem more valuable to 

 have continuous records of successive yields 

 of a single field. However, the yield of a 

 State is probably more reliable in showing 

 changes in productivity because it combines a 

 great number of individual fields and averages 

 the results of all the individual methods used 

 in the community. 



"When the yield per acre for a State or 

 country under practically constant acreage is 

 shown to have increased during a long period 

 of years we may assume that the productivity 

 has not declined. Much of this increase may 

 result from improved methods of agriculture, 

 better selection of seeds and crops, or the in- 

 troduction of more live stock; and, for the 

 purpose of our present inquiry, we need go 

 no further to explain the increase or main- 

 tenance of productivity. 



"There are two matters, however, which 



