212 
The Present State of Agriculture 
n ecllt to which his results will be entitled ; and sometimes a 
mere perusal of the remarks he appends to them explains suffi- 
ciently the knowledge and mental tendencies of the experimenter, 
and the amount of attention which his trials deserve. 
But without troubling you with any of the more refined modes 
of criticism, which when applied to a table of results often satisfy 
the reader that the whole must be unconditionally rejected, I will 
show you how the neglect of one or two very simple and intel- 
ligible precautions has rendered suspicious at least three-fourths 
of all the experiments hitherto published, and has necessarily 
excluded them from that body of satisfactory evidence on which 
safe opinions are to be based. 
First. In nearly all cases of experiments with what have been 
called specific manures, the limits of natural differences in the 
crop on different parts of the same field have not been previously 
or simultaneously determined. One portion of the field has been 
left undressed, and the produce of that one part has been taken 
as the standard, with which the produce of the dressed or manured 
parts was to be compared. But how incomplete this method is, 
and how unlikely to lead to the truth, the following examples 
will show you : — 
a. As to Turni})s. — In 1843 Mr. Dockar, of Findon farm, 
Aberdeenshire, made experiments on turnips, which were after- 
wards published ; and from two separate eighths of an acre of the 
same field, to which no manure had been added, he obtained 
First portion . . 8 tons 11 cwt. 
Second portion . . 6 IG „ 
Difference . . 1 „ 15 „ 
Or a difference of If ton upon a crop of 7 or 8 tons. 
A similar result was obtained at Ersklne, in Renfrewshire, 
where two portions of the same field yielded respectively 
First portion . . 12 tons 17 cwt. 
Second portion . .11,, 8 ,, 
Being a difference of . 1 „ 9 „ 
These two differences are equal to 25 and to 12 J per cent, of 
the whole crops respectively- 
h. So in the case oi natural hay. Mr. Chalmers of Monkshill 
found two several unmanured eighths of an acre of an experi- 
mental field, to yield respectively at the rate of 385 and 281 
stones of hay per acre, being a difference of 104 stones, or nearly 
two-fifths of the whole. 
I do not quote more exam])les. These are enough, I think, to 
entitle us to ask of experimenters — Are such differences in the 
natural produce of different parts of the same field common, or 
