108 
INVESTIGATIONS ON KOTHAMSTED SOILS. 
During the last six years of the period of fifty years which we are 
considering plat 2b yielded 40J bushels of grain per annum and 38f 
hundredweight of straw, while plat 7 (400 pounds of ammonium 
salts, with full minerals) averaged only 34f bushels of wheat and 33| 
hundredweight of straw. During the same six years plat 2a aver- 
aged 30J bushels of wheat and 28f hundredweight of straw. Clearly, 
therefore, plat 2b has reached, as would be expected, a state of high 
fertility, while plat 2a, though at the time of the last soil sampling' it 
had for nine years received the same treatment as 2b, was neverthe- 
less far behind it in fertility, the previous forty years of dung still 
telling very markedly on the latter plat. It is much more difficult to 
estimate the yearly addition of phosphoric acid to the dunged plats 
than to estimate it in the case of the chemically manured plats, for 
the phosphatic manures are of fairly definite and uniform composi- 
tion, while dung is necessarily variable and is very difficult to sample 
so as to fairly represent an annual application of 14 tons per acre. 
An}^ estimate is therefore, at the best, to be regarded as only an 
approximation. Sir Henry Gilbert estimates that the dung applied 
contains 0.25 per cent of phosphoric acid, and since this small per- 
centage represents in fifty years 3,920 pounds per acre, it is clear 
that there is room for considerable error. If the estimate of 0.25 
were 0.05 too high or too low as compared with the real average per- 
centage in a material of so indefinite and so fluctuating a composition 
as farmyard manure, the error of estimate on the fifty years would 
amount to nearly 800 pounds. 
In the following table the estimate of 0.25 per cent is, however, 
taken. 
Plat 3 (continuously unmanured) is again taken as a standard, the 
sum of the estimated loss of phosporic acid per acre on plat 3 and of 
the estimated gain on plat 2b representing the estimated excess per 
acre in the latter plat. 
Table 59,—Broadbalk wheat soils, samples collected in October, lS93—Plat .'b 
manured with 14 tons farmyard manure per acre per annum for fifty years. 
Pounds. 
Estimated excess of total phosphoric acid per acre over plat 3, 
at the end of fifty years - - 3, 086 
Excess of total phosphoric acid per acre over that in plat 3, 
found by analysis in first ( .) inches of soil 2, 062 
Excess of phosphoric acid per acre (as compared with plat 3) 
soluble in 1 per cent citric-acid solution, found in — 
First !) inches. 1,105 
Second iJ inches 141 
Third 9 inches - 36 
Total (27 inches) ... 1,282 
Ii is evident tbat, if the estimated quantity of phosphoric acid 
applied in the farmyard manure throughout the period of fifty yeanl 
