64 THE FORM OR MODE OF APPLICATION. 



was made ; and with the produce of its own undressed plot, the 

 results of each series were compared. But the two undressed 

 portions gave results differing as much as six bushels from each 

 other a difference which might be due to natural differences 

 in the soil ? but which Mr Main thus explains : 



" A stiffish breeze was blowing at the time of the top-dress- 

 ing ; and as the undressed portion was chosen in the vicinity of 

 that top-dressed with Peruvian guano, the dust from this sub- 

 stance was blown upon it. The guano plot was thus injured ; 

 and there can be no doubt that, in proportion as it was injured, 

 the undressed portion was benefited."* 



If this explanation be correct, then there is no standard with 

 which to compare the other results in the same series of ex- 

 periments 5 and being incorrect, the apparent produce of the 

 undressed portion ought not to have been published, as it is only 

 fitted to mislead. If, on the other hand, the difference be caused 

 by natural differences in the soil and, on a crop of 64 to 70 

 bushels, a difference of 6 bushels is not very great there then 

 is no certain standard with which to compare either of the series 

 of results obtained. The several results may be compared among 

 themselves ; but they do not enable us to judge of the absolute 

 effect of any of the substances employed. 



The mode in which a substance is applied, therefore, should 

 be such as to exclude all causes of doubt ; while duplicate or 

 triplicate experiments in every case should enable us to elimi- 

 nate such causes when they accidentally intervene. 



3. The form in which the substance is applied, has also 

 much to do with its apparent success. This has lately become 

 more generally understood, from the economical benefits which 

 have in so many cases followed from the practice of dissolving 

 bones by means of sulphuric and other acids. 



It may be laid down as a general rule, that the more minutely 

 any substance is divided, the more immediate and the more 

 sensible will be its effect upon a given soil or plant. The 

 reason of this is, that not only is a substance, when thus divided, 

 more widely and equably diffused over a field, and more gene- 

 rally brought within reach of the roots, but it is also more 

 * Transactions of the Highland Society, Jan. 1848, p. 157. 



