LOSSES OF MOISTURE AND PLANT FOOD BY PERCOLATION. 31 



The saving by cultivation is 3.52 inches. The saving due to the 

 manure would suffice for the production of the following crops, if it 

 could all be used: 



Manure, October 15. 



Cotton, pounds 1(57 



Corn, bushels 13 



Wheat, bushels 8 



Oats, bushels 19 



Hay, pounds 1600 



With the clays and heavy loams, the saving is much less. 



The important effect of manure on the saving of moisture with the 

 eanrls and sandy loam soils deserves especial emphasis. Manure or 

 green crops are undoubtedly needed by these soils, not only for this pur- 

 pose, but for the other favorable effects upon the condition of the soil. 

 But it appears that an application of ten tons of manure per year on the 

 sands and sandy loam soils may result in. a saving of moisture from loss 

 by evaporation sufficient to make more than an average crop of cotton, 

 and nearly an average crop of corn. 



PERCOLATION OE NITRATES. 



The plan of the work at first included only the estimation of the 

 nitrates from the pots not cultivated, which received no addition, and 

 which received manure March 15 and October 15. Later, however, 

 the plan was expanded to include the estimation of nitrates in all the 

 percolates. 



The nitrates were in each case estimated colorimetrically by the 

 phenol -sulphuric acid method the same day that the percolates were 

 collected. 



We will discuss first the results from the work which was carried 

 out on the original plan, and take up the other nitrate work on another 

 page. 



Table No. 13 shews the nitric nitrogen in the percolates for 1911, 

 1912 and 1913 in parts per million of the percolating water. This gives 

 the concentration of nitric nitrogen in the percolate. 



Table No. 14 shows the quantity of nitric nitrogen lost. per pot per 

 month, in the groups studied. 



As pointed out elsewhere in this Bulletin, a heavy rain in December, 

 1913, exceeded both the capacity of the free space of the pots to retain 

 on the surface, and the capacity of the bottles which received the per- 

 colates,, and, for this reason, December is not included in the work for 

 the year. As, however, the analyses are of some interest, they are re- 

 corded in the- tables. 



