Price of Wheat, over 40 Harvest-Years , 1852-3 to 1891-2. 95 
seasons, we have adopted an estimate rather higher or rather 
lower, as the case may be, than the exact experimental results 
would indicate. Accordingly, the figures show that, over the 
eight years 1860-1 to 1867-8, which included several of very 
high productiveness, the average yield on the selected plots was 
29 bushels, whilst the average of the annually adopted estimates 
for the United Kingdom over the same period was 28| bushels, 
or rather less. Again, over the eight years 1876-7 to 1883-4, 
which included some exceptionally bad seasons, the selected 
plots gave an average of only 23 bushels, whilst the average of 
the annually adopted estimates of the yield of the United Kingdom 
was 24| bushels. Taken over the whole period of the thirty- 
seven years of normal treatment of the land, however, the 
selected plots indicate an average of 27 bushels, against 27^-, 
according to the annually adopted estimates. In other words, 
there is, over the thirty-seven years, only a difference of a 
quarter of a bushel per acre per annum between the direct 
indication of the produce of the selected experimental plots and 
the results of the annually adopted estimates. Lastly, even in- 
cluding the abnormal results of the last three years, the average 
for the forty years is seen to be 27^ bushels according to the 
actual results of the selected plots, and 27^ bushels according 
to the annually adopted estimates, showing, therefore, a difference 
of only \ bushel over the forty years. 
We will now compare the annually adopted estimates of the 
produce of the country at large with the results given in the 
last column of the Table, which are obtained by deducting the 
subsequently recorded net imports from the estimated total re- 
quirement for consumption, each year adding the amounts required 
for seed, and dividing the product by the number of acres under 
the crop. It is seen that, over each of the first four eight-yearly 
periods, there is a difference of only ^ bushel per acre between 
the average of the annually adopted estimates and the average 
of the results founded as above on consumption and imports 
over the same periods. 
Over the subsequent five years, however, whilst the average 
of the annually adopted estimates indicates 28$ bushels per acre 
for the home crop, the calculated result founded on consumption 
and imports shows 30f bushels, or a difference of 1$ bushels in 
excess of the average of the estimates. On the other hand, over 
the last three years, whilst the average of the annually adopted 
estimates is 30£ bushels, the result founded on consumption and 
imports indicates only 26$ bushels, or a deficiency of 4$ bushels. 
The facts are, as will be seen farther on, that the amounts 
available for consumption per head of the population from home 
