in its Relations to Chemistry and Gcohr/ij. 
211 
Ami yet from the mass of published experiments, and from the 
opinions — often hasty and badly founded, sometimes extravagant, 
and almost always far too general — which have been based upon 
them, there is much chatt" to be winnowed, and many hurtful 
seeds to be sifted out. And that it should be so is not surprising, 
when you consider the numerous difficulties that lie in the way of 
the practical farmer in attempting to make experimejits with 
scientific accuracy. He is supplied, perhaps, with impure 
materials, which he cannot himself examine, and lor admixtures 
or adulterations in which, he can consc([uently make no allow- 
ance in judging of his results. Again, the carelessness, the 
culpable and generally concealed neglect, and even, in some 
cases, the opposition and unwillingness of his servants, are diff- 
culties which he can scarcely altogether overcome: and further, 
he is himself most likely inexperienced in accurate observation, 
and thus overlooks some necessary precaution in conducting his 
experiments, Mliich renders their results less trustworthy. 
It has too often hapj)ened, also, that experiments have been 
undertaken without a definite purpose. The trials have been 
made without a clear knowledge of what was to be looked for, 
and points have thus been passed over which it was of import- 
ance to observe. The old alchymists, who made experiments 
upon every thing that came in their way, put materials of which 
they knew little into their retorts and crucibles, and heated them 
in their furnaces, on the chance of something curious or valuable 
turning up. The modern chemist, on the contrary, selects 
materials of known composition, mixes them in known propor- 
tions, heats them to a definite temperature, in a known way, and 
looks for a known or expected result. And so conducted, and 
carefully watched, must our future field-experiments be, if they 
are to advance or widen our real knowledge. 
From the obstacles above noted haTe arisen, in many cases, the 
contradictory results we so frequently observe among our published 
experiments; and, generally, no means are afforded us for deter- 
mining upon which of two or more discordant statements the 
greater reliance is to be placed. The soil, or the substance 
applied to it, has not been analysed — the history of the past 
treatment and present ctmdition of the land is not given — the 
peculiarities of ihe weather, or of the local climate, are not noted 
— the physical geography of the neighbourhood is left out of 
view — the character, willing and attentive, or otherwise, of the 
workmen is not stated — or we know nothing of the general 
habits, skill, and education of the person who j)ublishes and is 
responsible for the accuracy of the whole results. 
Sometimes a previous acquaintance with the individual enables 
us beforehand to form a general opinion as to the degree of 
p 2 
