234 



and lime, assuming the difference in yield to be due to tbe fertilizers 

 api^liedj and finally the "value of the net increment due to fertilizers," 

 and the financial results. 



The two chief difficulties experienced were richness, and unevenness 

 in fertility of the soil. The effects of disturbing- conditions, including 

 " the remains of former manure piles," were such that in only two cases 

 was the laud " tolerably even in fertility." This unevenness is strik- 

 ingly illustrated by the yields of the un manured plats in the same field. 

 Thus on one field, a fine gravelly loam, the greater part of which had 

 been in grass and unmanured for ten years (though ^' a small portion 

 had been cultivated within five years, receiving but little manure, and 

 since that time had been in grass and unmanured"), the yield of corn 

 on the five unfertilized plats was at the rate of 19.8^ 11.5, 16.9, 23.2, and 

 8.7 bushels per acre, respectively; and in another instance, on a soil 

 "composed of very fiue sand," which " had never received much manure, 

 was cultivated last seven years ago and had subsequently been used 

 as a pasture," the unmanured plats yielded, respectively, at the rate of 

 49.6, 41.8, 49.1, 27.6, and 32.1 bushels per acre. 



That the eight different soils tested differed very widely from one 

 another in natural fertility is evident from the fact that the average 

 yields of shelled corn on the five unmanured plats of each experiment, 

 calculated for 1 acre, varied from 6 to 50 bushels, and of stover from 

 ojO to 2,836 pounds, the average yield on the 40 unfertilized plats in 

 the trial being 26.3 bushels of shelled corn per acre. 



The results seem to indisate that on three of the soils tested, potash 

 was the element most needed, and on one phosphoric acid, while in four 

 cases the results were not pronounced enough to be conclusive; in 

 every case except one tbe increase of shelled corn over the jield from 

 unmanured plats was larger with barnyard manure than with the com- 

 plete fertilizer, aiid in six instances this increase was larger with barn- 

 yard manure than with any other fertilizing material. 



It is evident that the indecisive results of several of these experi- 

 ments were due to the character of the soils rather than to any defects 

 in either the planning or executing of the work. The following general 

 conclusions are warranted by the details given. 



(1) The results of our experimeuts l)riug out in a striking manner the fact that 

 soils vary widely in their requirements ; * * * that results obtained in one local- 

 ity may be inapplicable in another j and appear to establish the wisdom of the policy 

 of local soil tests. 



(2) It is evideut that only when the farmer knows what his soil requires can he 

 produce the best economical results. It is folly to continue the indiscriminate and 

 blind use of fertilizers. 



(3) Tbe best method of ascertaining what is needed in any given case to produce a 

 particular croj) is to put the question to the soil itself; and this method, though re- 

 quiring care at all points and cautiou in forming conclusions, is not, in reality, diffi- 

 cult. Such experiments should abundantly repay the investigator in the practical 

 money value of the results. 



