The first step in answering this is to determine just what 

 kind of plant food has been supplied in each case, and the pro- 

 portion of the several kinds. If we take the fertilizers used on 

 those plots which gave the three highest yields in each set, and 

 average the per cent, of nitrogen, phosphoric acid and potash, 

 we get the results given in table 4. 



Table 4. Composition of best Chemicals used. 



The upper half of this table shows what kind of a fertilizer 

 proved best for the production of corn on each of the seven 

 farms, the last column averages all of these. The lower half of 

 the table shows the same thing for the production of fodder, 

 consequently these results may be regarded as applicable where 

 the design is to raise ensilage; combining these averages and we 

 may fairly claim, so far as the teachings of these experiments 

 are concerned, that the best results come from a fertilizer, 

 with the following chemical composition: (beside it is given 

 the average analysis of 18 fertilizers sold in N. H. in 1889) 



Chemicals Average of fertilizers sold 



producing best results. in N. H. in 1889. 



Phosphoric acid 9.25 11.08 



Potash 1 1.3 2.57 



Nitrogen 3.5 2.45 



The difference is very easily seen and we are forced to 

 conclude that our prepared fertilizers are deficient in potash. 



Or if we select from table 2 those three plots which yield 

 highest in the average of all of the sets namely, 9, 13 and *6 

 the average composition is as below. 



Husked Corn. Fodder, 



Phosphoric acid 10.7 7. 



Potash 9.5 23.1 



Nitrogen 2. 2.1 



The best plots on fodder were 13, 9, 5. 



Average of the two. 



8.8 



16.3 



2.0 



*6 received a variety of fertilizers among the different experimenters. 



9 



