2 
MANURING EXPERIMENTS WITH PADDY RICE. 
of each nutrient and factor of production as will amply suffice 
for a maximum yield, excepting, however, the nutrient under 
experiment which must be applied in different doses. If the 
soil is originally poor in this nutrient, the yield will be, up to 
a certain limit, proportional to the amount of the nutrient 
given ; and on the other side, if it be a priori rich in it, the 
yield will not be much influenced by different doses of that 
nutrient. The upper limit to which the yield can be econ¬ 
omically increased, gives a reliable measure for the quantity 
of the nutrient to be applied in practice. 
Among the important vegetable nutrients, nitrogen and phos¬ 
phoric acid exist in the manures in very different forms , some 
of which can be directly consumed by the plants, and conse¬ 
quently have a speedy effect on the crop ; while others must 
be decomposed in the soil before plants can take up their 
nutritive ingredients, which circumstance may retard and 
diminish the effect. 1 Comparative trials with various forms 
of nutrients accordingly constitute also an essential object of 
agricultural experimental work and are just as necessary as 
researches on the quantities of available nutrients needed for 
a maximum yield of a special crop. 
Manuring experiments which are to decide upon subjects of 
general importance, cannot be well carried out on large plots, 
because of the frequent inequalities of the soil and difficulties 
of making the general factors of production quite equal on all 
plots. Small patches, on the other hand, admit of easy and 
accurate management and allow of constant control over all 
the factors that influence the development of plants. The 
only draw-back to trials on a small scale is that they are in a 
more favourable condition of light and heat than large plots, 
but in many cases this source of error can be avoided by 
surrounding the small plots with the same kind of plants, so 
as to confer on all trials the conditions of a single large field. 
On the principles explained in the preceding lines, and 
i See Bulletin No. 6, p. 25. 
