in its Relation.'! to Chemistry and Geology. 
221 
Third. — I have dwelt thus \ov\^ on the experimental and di- 
reetly })vactical part of our progress, as it is the one which most 
plainly hears upon the profitable prosecution of the agricultural 
art. But it is, in realit_y, no less important to inquire what 
progress we have made in correcting our theoretical opinions, 
and in fixing upon a surer basis the explanations of rural pheno- 
mena which we are called upon to give. That our position in 
regard to agricultural theory has considerably improved during 
the last five years will be made plain to you by a few examples. 
Thus — 
a. It is known that the bodies of animals contain much nitro- 
gen. This nitrogen they obtain from their food, and this food is 
all of vegetable origin ; for though man or the lion may devour 
the ox, the sheep, the goat, &c., yet these animals live solely on 
the herbage which springs from the soil. Now vegetable food is 
believed to be more valuable for the production and support of 
those parts of animals which contain nitrogen, in proportion as 
the per-centage of nitrogen present in itself is greater. It comes 
nearer to the composition of the animal parts themselves. It has 
also been long received as a practical truth that the proportion of 
nitrogen in wheat and other kinds of grain was very much under 
the control of the cultivator — that by a proper adjustment of the 
quantity and quality of the manure he could halve or double, at 
pleasure, the per-cenlage of nitrogen contained in his crops of 
corn. Numerous experiments and analyses, supposed to be care- 
fully made, and therefore satisfactory, appeared to have proved 
this ; and you may have heard this fact urged by less cautious 
persons as a proof of the dominion which chemical research has 
already given us over the operations of vegetable nature. 
But this opinion has not borne the test of rigid experimental 
and chemical criticism. You will find doubts thrown upon it in 
my earliest agricultural writings; and experiments and analyses 
published by myself, by Schlossberger, and by Mr. Lawes, ap- 
pear to justify us in banishing it altogether from our books. The 
correction in this case has arisen not so much from better-made 
field experiments as from more skilfully conducted analyses. 
After the experimental crops are gathered they must be ana- 
lysed, and it was incorrect analysis which gave countenance to 
the opinion I have mentioned. If you consider how much the 
mixture, and above all, the skill and good faith of the manufacturer. 
Without a knowledge of all this the chemist cannot fairly judge how far 
the sample he has examined truly represents the whole, and wliat proba- 
bilily there is that the manure, made at different times, will possess a con- 
stant and uniform composition, and can therefore be safely recommended 
to the rent-paying farmer. 
