On the Rain and Drainage - Waters at Rothamsted. 
25 
plots receiving potash an average of 4 2 parts of potash per 
million. The former quantity would correspond, on the pre- 
vious assumption, to a loss of 3'6 lbs., and the latter to 9"5 lbs. 
per acre per annum. As to the probable correctness of calcu- 
lations based on this assumed amount of drainage we shall speak 
in the concluding section of the paper. In the absence of drain- 
pipes, a part of both the phosphoric acid and potash in the 
drainage-water would doubtless be retained by the subsoil. 
The figures showing the amount of soda in the drainage- 
waters are much more regular than those for potash or phos- 
phoric acid. The water from six plots receiving no soda 
contains a mean of 6'1 of soda per million. The water from 
five plots receiving 100 lbs. of sulphate of sodium annually 
contains a mean of 11'6 per million of soda. Where 366i lbs. 
of sulphate of sodium are applied (Plot 12), the soda becomes 
24"6 per million. Where 550 lbs. of nitrate of sodium are 
applied (Plot 9), the soda in the drainage is increased to 56" 1 per 
million. As sodium-salts are very soluble, and the retentive 
power of soil for soda is very small, it is probable that with a 
regular supply of soda there will be a regular discharge of a 
nearly equal amount in the drainage-water. The quantity of 
soda annually assimilated by the wheat-crop is very small ; on 
three only of the plots in Broadbalk does it exceed 1 lb. 
per acre. 
The soda found in the drainage-water from the unmanured 
plots will be derived from rain ; and to a smaller extent from 
diffusion from the neighbouring: soil or subsoil. If we assume 
the chlorides present in the rain at Rothamsted to exist as 
common salt, the rain analysed during forty-three months, in 
1877-80, would have supplied annually about 11*7 lbs. of soda. 
A part of the sulphates in rain may also exist as sodium-salts. 
It is obvious that if the phosphoric acid and potash applied in 
manure have not escaped to any serious extent in the drainage- 
water, the portion unused by the crop must remain stored 
up in the soil. In October 1865, samples of soil were carefully 
taken from eleven of the plots in Broadbalk, representing the 
first, second, and third 9 inches from the surface. The soils 
from Plots 2, 3, ba, la, and lOa, were examined by Hermann 
von Liebig.* He determined the substances soluble in boiling 
dilute acetic acid, and the amount of phosphoric acid soluble in 
dilute nitric acid. Assuming that all the soils were originally 
alike, and knowing the substances applied as manure, and 
removed in the crop from each plot, it becomes possible to 
* " Bodenstatik und Bodenanalysen." ' Zeitsrluift des JandwirthscLaftliclien 
Vereincs,' 1872. Also Jour. Chcm. Soe. 1872, 318, 837. 
