ill its Relations to Chemist) 1/ and GeoUxji). 
215 
Anil on the field on the Kerrytonlia farm in Bute, on which 
the experiment already quoted was made, 
Tons. Cwt. 
24 carts of dung and 16 bush, of bones in one experiment gave 27 7 
24 „ 14 „ „ 29 7 
Difference . • • . • .20 
Again, on the same iarui. 
Tons. Cwt. 
24 carts of clung and 3 cwt. African guano gave in one ex- 
periment . . . . . . . . 25 4 
24 carts of dung and 2\ cwt. guano . . . . 28 7 
Difference 3 3 
Are these differences to be ascribed to the errors or careless- 
ness of the experimenters or their servants ? or do they signify 
that no exact numerical results are ever to be obtained from such 
experiments? It is unnecessary to discuss these questions in this 
place. It is enough for my present purpose to observe, that, inas- 
much as the tables of results which we at present possess exhibit 
for the most part only one illustration of the effect of each sub- 
stance or mixture employed, no strong reliance can be placed 
upon them as indications of what their mean or average effect 
would be upon the field as a whole, or upon its several parts 
taken singly. 
I do not dwell on the very small portions of land on which 
some have made their experiments — on the form of these por- 
tions, as when one or two drills of a field are selected — on the 
want of attention to the previous chemical treatment, the his- 
torical husbandry followed upon it, and the chemical condition of 
the land at the time — for all of which, if we knew them, allovv- 
ances ought to be made in the after criticism to which the results 
are subjected. 
But the application of these and other tests has compelled me 
to the conclusion that a ver}- small number indeed, out of the host 
of experiments made during the last five years, is really deserving 
of more than a general credit. They are to be trusted, as showing 
that this or that substance adds to the verdure, bulk, or weight of 
this or that crop in certain circumstances ; but they are not to be 
depended upon as to how much these several substances do so 
either relatively or absolutely, or as to the degree by which their 
action is affected by this or that circumstance. In other words, 
the application of weight and measure in its strictest sense to this 
branch of applied science has yet to be made ; and the field of 
precise agricultural experiment, instead of being pretty well colo- 
nized, as some think, may be said to lie still open and unoccupied 
before us. 
