THE HBOADBALK WHEAT SOILS. 19 
ley soils in Hons field. I was much struct with the great variations 
shown by the subsoils of tin* wheat plats in their lowermost depths. 
The lower subsoils, at corresponding depths, were in some cases 
mere chalk, in others mixtures of clay and chalk, and in Others of 
chalk, clay, and gravel; and this lack of any kind of apparent uni- 
formity ><> low down made it obvious that the lower samples were 
likely to be of very little value for the purpose of comparison of 
their fixed mineral contents or of their organic nitrogen. My mind 
was probably mainly concentrated upon the phosphoric acid and pot- 
ash problems which I was myself about to be allowed to investi- 
gate, and it appeared to me that from thai point of view there was 
not much interest in the samples below the second or third depth. 
Sir John Lawes pointed out that this would not apply in anything 
kike the same degree to diffusible constituents, e. g., nitrates. It 
appeared to me, however, to be very unlikely that even the migratory 
constituents would present much interest al a depth of 71 feet, for at 
thai depth, 01 even considerably short of it, it seemed natural to sup- 
pose that the subsoil contents washed down from the surface would 
have not merely descended, but would have also diffused laterally to 
such an increasing extent that the subsoils would surely have their 
proper soluble contents merged into one another; for the wheat plats, 
although something like half an acre in area, are arranged in long 
strips parallel to one another, and not in squares. This view I was 
rash enough to express pretty confidently, witli the result that Sir 
John Lawes —who, I believe, inwardly felt a like conviction — decided 
not to spend further time and labor incompleting the sampling of the 
deeper subsoils, especially as the season was getting late and he was 
anxious to get the wheat sown. Now it happens that my view as to 
the lateral diffusibility of the nitrates, even at so great a depth as 7 A 
feet, was incorrect, their downward settlement being much more ver- 
tical than I had anticipated; and the results of subsequent analyses 
of even the deepest subsoils by sir Henry Gilbert were such as to 
make it a matter of regret that the original plan of sampling to 90 
inches throughout the whole series of plats was not carried out. But 
for my unlucky visit, Sir John — even, perhaps, a little in the face of 
his own inclinat ion -would have completed the sampling plan that he 
had prearranged with Sir Henry Gilbert, who, I am afraid, does not to 
this day quite forgive me. I now publicly express my sorrow for the 
gaps which, through my fault, will be found in certain interesting 
tables which we shall have to deal with by-and-by. 
