750 



should all be exactly of the same size. All the plots are to be 

 ploughed in the one day, and also sown and manured on the same day. 

 If we take, say, wheat as the test crop, each plot must be sown 

 with the same quantity and same variety of wheat. No. I receives 

 no manure ; No. 2 receives one-tenth of what is required for an acre, 

 of normal manure No. I ; No. 3 plot receives the same manure, with 

 the nitrogen omitted ; No. 4 receives the same manure without the 

 phosphates ; No. 5 plot receives the same manure without the 

 potash ; No. 6 plot no manure. The crop from each plot should be 

 gathered the same day, and the grain and straw weighed separately. 

 This completes the analysis of the soil, and we have only to calculate 

 the results from the figures obtained ; and to show how this is done 

 we will take a series of experiments made by G. Ville, at Vincennes, 

 on wheat, with the following results : 



Crop per acre, bushels. 



No. 2 plot, normal manure ... 43 



,, 3 ,, manure without nitrogen 14 



4 M n phosphates 26^ 



5 ,, ,, potash ... 31 



,, i and 6 plots, no manure, averaged 12 



The conclusions arc evident. This soil requires, above all, 

 nitrogenous matter, it is also insufficiently supplied with potash and 

 phosphates of lime. From the above results \vc arc in a position to 

 say what quantity and class of manure is required tor any ^ivcn 

 crop. Hut we have still another tiling to consider, how to maintain 

 the soil in a fertile condition, and to know when it is ^cUin^ 

 deficient in any of the plant foods. 



To accomplish this the amounts of phosphoric acid, potash, and 

 nitrogen added in manure from the commencement of the year in 

 which the test has been made, must be kept account of in a book 

 for the purpose and for every succeeding year, thus, the amounts 

 per acre, were, say : 



