On the Rain and Drainage - Waters at Rotliamsted. 
13 
collections, namely those on Jan. 5, May 18, and Oct. 26, 1872 ; 
and on Jan. 19 and Feb. 26, 1873, included drainage-waters 
from nearly all the plots. The collections represented both large 
and small runnings of the drains. In Table XLI. will be found 
analyses of the waters collected on Jan. 5 and on May 18, 1872. 
The first of these collections represents the first general running 
of the drains since the application of the autumn manures ; the 
pipes were in moderate flow. The second series of waters 
represents a small flow of the pipes later in the same season, and 
after the application of nitrate of sodium to Plot 9. The mean 
composition of the five general collections will also be found in the 
table. * The whole of the results have been published in detail 
in the ' Sixth Report of the Rivers' Pollution Commission,' 
1874, pp. 58-68. 
Since the examination by Dr. Frankland, the composition of 
the drainasre-waters has been further investigated at Rotham- 
sted. The analyses previously made had shown that very con- 
siderable quantities of nitric acid were removed in the 
drainage-waters from the plots receiving ammonium-salts or 
nitrate of sodium. This loss of nitrogen by drainage appeared 
to be, from an agricultural point of view, the part of the sub- 
ject most requiring a fuller investigation ; the work done at 
Rothamsted has therefore been chiefly directed to this part of 
the question. The examination of the waters has consisted in 
the determination of the quantity of nitric acid and chlorine 
present, with qualitative testing for ammonia and nitrous acid. 
The methods employed were the same as those already men- 
tioned when speaking of the analysis of the waters from the 
drain-gauges. 
Bv confining- the examination of the waters to the determina- 
tion of the two constituents just named, it has been possible to 
examine a far larger number of waters than could otherwise have 
been accomplished. The analyses made include one series of 
waters collected in 1876, eight series collected in 1877-8, 
thirty-six series collected in 1878-9, fourteen collected in 
1879-80, and forty collected in 1880-1. During the last three 
seasons, individual or mixed samples, including nearly every 
running of the drains, have been submitted to analysis. The 
total number of samples analysed is about 1300. The greater 
part of these analyses were made by Mr. W. H. A. Peake. We 
shall be unable to give the results in detail from want of space. 
In considering the facts which the investigations of Voelcker, 
* The mean composition of the water from Plots 2 and 9, is calculated from 
four analyses, that of Plot 16 from three analyses. The organic carbon and 
nitrogen in the water of Plot 10 is the mean of four analyses. The other means 
represent five analyses. 
