Section I. 
SOIL SAMPLING. 
The methods of soil sampling adopted at Rothamsted have been 
previously described to you, but it seems desirable, in a memoir which 
embodies such complete statistical results as will here be recorded, 
that the details relating to sampling should for purposes of reference 
be included. Indeed, it is desirable that this should be done, if only 
to again emphasize the general necessity of minute precautions and 
regular procedure in sampling any soils which either possess or may 
be likely to acquire an experimental history. Those who have the 
charge of experimental stations, and who may not have brought their 
minds to bear fully on the capital scientific importance of correct soil 
sampling, may derive useful hints from the description, while those 
who are already alive to the difficulties that may follow upon disre- 
gard of certain precautions will, from that very sense, bear with me. 
For, although attention has been directed very frequently to this 
matter, there is too much reason to suppose that not a little other- 
wise good soil work has been vitiated by reason of insufficient atten- 
tion to the drawing of representative samples. 
The first determinations of nitrogen in the Rothamsted soils were 
made by soda-lime combustion by the old platinum method, in the 
surface soil of the Broadbalk wheat field, as long ago as 1846; but it 
was soon recognized that results obtained on samples collected with- 
out careful attention to area and depth were of little value, and were, 
indeed, misleading. In 185G the method of collection then and sub- 
sequently adopted was devised, and I perhaps can not do better than 
quote directly from Professor Wanngton's lecture. 1 Professor War- 
Lngton says: 
A frame made of stout sheet iron, in shape a rectangular prism, open at lop and 
bottom, is driven into the soil by repeated blows of a wooden rammer, till the soil 
has the same level inside and outside the frame. The soil inside the frame is then 
cut out. and constitutes the sample of the first depth or surface soil. That the 
frame is accurately emptied is ascertained by trials with a wooden gauge of the 
sam*' depth as the iron frame, if a sample of the next depth is to be taken, the soil 
is cleared away around the outside of the frame till the level is reduced to that of 
the bottom of the frame, the frame is then driven down again, and the former 
operations are repeated. 
Soil sampling at Rothamsted is usually carried down to three depths, but in a 
good many cases it has been carried down to twelve depths. The area of the sam. 
1 U. S. Dept. Agr., Office of Experiment Stations Bui. 8, p. 89. 
