170 AGRICULTURE 



larly it might be argued that plot No. 1, 

 receiving no manure, is unnecessary ; but, 

 as the yield of this plot will prove interesting 

 as throwing light upon the natural capa- 

 bilities of the soil, it is perhaps on the whole 

 desirable to have an unmanured plot. This 

 simple five-plot test may be extended almost 

 indefinitely, and certainly in the majority 

 of cases it should be supplemented by 

 additional plots. For instance, it is probably 

 desirable to ascertain what is the profitable 

 limit of the use of a phosphatic manure, 

 and for this purpose a sixth plot could be 

 laid down, receiving 7| cwt. of superphosphate 

 along with the nitrate of soda and kainit of 

 plot 2. By so doing we have the opportunity 

 of ascertaining (a) the effects of the omission 

 altogether of superphosphate (compare plots 

 2 and 3), and (b) the effects of a moderate 

 as compared with a large dressing of super- 

 phosphate (compare plots 3, 2, and 6). Of 

 course a still larger dose of superphosphate 

 could be tried upon an additional plot, but, 

 for ordinary practical purposes, this is per- 

 haps scarcely necessary. It will, however, 

 be useful, in many cases, to ascertain what 

 amount of nitrate of soda and kainit can be 

 profitably employed, and for this purpose 

 we can add the following plots : 



