On the Rain and Drainage - Waters at Rotkamsted. 337 
position of the organic matter corresponds with the second oi 
the above ratios. The gradual increase in the proportion ot 
nitrogen contained bj organic matter as oxidation in the soil 
proceeds, is strikingly shown by the determinations of carbon 
and nitrogen which have been made in the Rothamsted soils. In 
the surface soil (first 9 inches) of pasture land, with roots as far 
as possible removed, the proportion of nitrogen to carbon is 
about 1 : 13. In the clay subsoil of the same land (fifth and 
sixth 9 inches) the proportion is about 1:6. In the soluble 
organic matter contained in drainage-water we have just seen that 
the proportion reaches 1 : 2*6. Respecting the nature of these 
nitrogenous organic bodies, and the part they possibly play in 
plant nutrition, very little is at present known. 
The chlorides found in the first two analyses greatly exceed 
in amount any quantity subsequently found ; this large propor- 
tion of chlorides was probably due to the previous manuring 
with guano. The amount of lime present in the waters was 
considerable, as shown by the "hardness." In the later analyses 
the total solid matter is chiefly made up of calcium salts- 
nitrate, sulphate, and carbonate ; alkali salts must, however, have 
been also present in considerable quantity. In the earlier 
samples, the alkali salts form the largest ingredient ; at least the 
calcium salts can account for only a small part of the solid 
matter. 
If we compare together the drainage-waters from the three 
gauges, we see that as far as the total solid matter and nitrates 
are concerned, the drainage from the 40-inch gauge is weaker 
than from either of the others, the order of strength is in fact 
20, 60, 40. This comparatively low proportion of nitrates in 
the drainage from the 40-inch gauge is shown in the first 
analysis made in 1870, and is equally shown in nearly all the 
analyses that have been made since ; the cause must apparently 
be sought in some original difference in the soil forming this gauge. 
We pass now to the more recent analyses of these drainage- 
waters. Since September, 1874, a sample has been taken from 
the drainage of each day, and mixed monthly samples prepared 
representing the drainage from each gauge. It being uncertain 
whether the samples could be analysed, the sampling was at first 
roughly done ; but since May, 1877, a fixed fraction of each 
day's running has been carefully taken, so that the mixed 
monthly sample might exactly represent the whole drainage. 
The earlier samples were not examined till the spring of 1877, 
owing to the lack of analytical assistance in the Laboratory. At 
that time many of the samples were found to be ill preserved, 
and had to be discarded ; determinations of nitric acid were 
then made in the remainder. Since May, 1877, determinations 
