33G 
On the Fain an I Drainaye -Waters at Rothamsted. 
The fifth collection, April 2-30, 1874, was a mixture of all the 
runnings during one month. Drainage was pretty continuous 
throughout the month, though at a very gentle rate, barely one- 
third of an inch having been collected from the 20-inch gauge. 
The preceding winter had been dry. The water from the 20-inch 
gauge was slightly turbid ; the others were clear. 
The first three collections are thus autumn drainage-waters, 
the preceding summers having been dry. Nitrification had 
doubtless been active in the upper layer of the soils during the 
summer months ; but as very little drainage had occurred the 
nitrates produced had not been to any considerable extent re- 
moved by rain. These drainage-waters were consequently all 
of a concentrated character, and particularly rich in nitrates. 
The fourth collection fell on the other hand towards the close 
of a remarkably wet winter, when the soil had been washed by 
the percolation of a large amount of water. The drainage-water 
now was less concentrated than before, and was especially poor 
in nitrates. 
The fifth collection was in spring, at the end of a dry winter. 
The rate of drainage having been very slow, the water doubtless 
represented the general discharge of the soil, and owed but little 
to direct channel drainage. The water is seen to be much more 
concentrated than that obtained at the fourth collection, but does 
not equal in this respect the three autumn drainage- waters. 
Looking at the analyses generally, we see that ammonia is 
cither absent, or occurs in very small quantity. The amount of 
organic matter dissolved in the water is but small ; it is increased 
when the water is turbid ; it is in all cases highly nitrogenous. 
The mean ratio of organic niti ogen to carbon in the drainage- 
waters from the three gauges is 1 : 2"6, 1 : 2 "9, and 1 : 3"1, the 
proportion of carbon apparently increasing with the depth of the 
soil. This, however, can hardly be established as a fact from 
the few analyses now before us. The proportion of carbon 
is highest in the turbid waters ; the mean ratio of nitrogen to 
carbon in the six turbid waters being 1 : 3 3, and in the nine 
clear or slightly turbid waters 1 : 2'6. Turbidity in a drainage 
water is a sign that direct channel drainage has occurred, matter 
being brought immediately from the surface. 
Dr. E. J. Mills has already called attention (' Trans. Chem. 
Soc.,' 1878, p. 64) to the constancy of the relation between the 
nitrogen and carbon of the organic matter found in clear well- 
and drainage-waters. He considers that the slow oxidation 
which organic matter undergoes in a soil finally reduces all 
forms of organic matter to a few simple compounds, in which 
the carbon and nitrogen have the relation Cj., : N ,, Cy, : N4, or C12 : 
N5 ; in the drainage-waters we are now considering the com- 
