The Ilidorij of a Field newly laid down to Permanent Grass. 19 
such fine vegetable debris. Against such a supposition is the 
fact that determinations of carbon in the mixture of the six 
1879 and of the five 1888 soil samples show scarcely any in- 
creased relation of carbon to nitrogen at the later date, the 
amounts being ]1"73 carbon to 1 of nitrogen in 1879, and only 
11*77 in 1888. Compared with these amounts, as already stated, 
there are from 13 to 14 parts of carbon to 1 of nitrogen in the 
surface-soil of the old grass-land at Rothamsted. 
At any rate, the figures in Table VII. show that the average 
amounts of nitrogen per acre due to separated roots were 
rather less at the later date, being 88'3 lbs. in 1879, and only 
77"9 lbs. in 1888. These amounts are much less than were 
attributable to separated roots in the case of the old grass-land. 
Upon the whole, therefore, the evidence clearly indicates that 
there is much less accumulation of vegetable debris in the 
surface-soil of the new, than in that of the old grass-land. As 
to the actual amounts of nitrogen due to vegetable accumula- 
tion in the surface-soil, if the whole amount were distributed 
over the 33 years, it would represent less than 3 lbs. per acre 
per annum. But, obviously, a larger part of the earlier than of 
the later accumulations will have been decomposed, and will 
have contributed to the increase of nitrogen and carbon found 
in the fine soil. Hence, more than the above calculated average 
amount, both of vegetable debris and of nitrogen in it, will be 
due to the later years. Still, the amounts of nitrogen due to 
such accumulations of visible and separable vegetable matter 
will not add many lbs. per acre per annum to the amounts 
remaining and found in the fine soil itself. 
From the whole of the results there can be no doubt that 
there has been a considerable accumulation of nitrogen in the 
surface-soil during the formation of the meadow — amounting, in 
fact, to an average of nearly 52 lbs. per acre per annum over the 
last 23 years. The question arises — whence has this nitrogen 
been derived ? 
In Table III. it was shown that, on the assumption of the lower 
percentage of nitrogen (0'64) in the dung applied, there were 
7'9 lbs. more nitrogen annually supplied in manure than were 
removed in the crops, over the 23 years 1866 to 1888 ; but that, 
on the assumption of the higher percentage of nitrogen in the 
dung (0-80), there were about 20 lbs. (20'1) more annually 
supplied than were removed in the crops ; or, even if we were to 
suppose that the dung contained TO per cent, of nitrogen, the 
excess in manure over that in the crops would only amount to 
35-4 lbs. per acre per annum. 
It is obvious that, even supposing there were no loss of the 
c'2 
