E. J. HUSSELL AND A. AprLBYAKD 



Broadbalk wheat plots, 1915 



401 



* The nitrogen in the crop is obtained as follows : 

 Unmanured plot, per acre 



Dunged plot, per acre 



In the preceding discussion we have regarded the curves as pro- 

 duction curves. They are not, however, entirely production curves, 

 because losses are always going on ; they only represent the balance 

 between gains and losses. The upward parts of the curve show that 

 production is predominating, but it does not necessarily represent the 

 whole of the production ; there is always the possibility of loss. The 

 downward portions of the curve show that the losses are predominating, 

 but, again, they do not necessarily show the whole loss, because there 

 may be some production all the time. The curves, therefore, tend to 

 be flatter than if they were solely production curves. 



We must now discuss the loss factors to see what light they throw 

 on the curves. Two causes are known for the loss of nitrate — ^leaching, 

 and absorption by the crop. On the fallow ground there is no crop, 

 and leaching is the only known cause at work. It remains to discover 

 whether this is sufficient to account for all the observed losses. 



The kinks in the nitrate curve for the dunged fallow plot during May 

 and at the end of July and the beginning of August correspond with 

 periods at high percolation as shown by the drain gauges (Fig. 9). It is 

 not necessary to invoke any other cause to account for these, and right up 



