Tlie History of a Field newly laid down to Permanent Grass. 15 
But with the higher percentage the weight of fine dry soil per 
acre was only 1,981,924 lbs., whilst with the lower it was 
2,108,068 lbs. ; so that there is only a difference of from 4,186 
to 4,094, or 92 lbs., in the calculated quantity of nitrogen per 
acre in the two cases. There can be little doubt that the lower 
percentage in the sample of greater actual weight of fine dry 
soil was due partly, at any rate, to the inclusion of a greater 
quantity of subsoil which would contain a lower percentage ot 
nitrogen. The percentage of nitrogen according to the mean of 
the determinations on the six individual samples is 0-2024, and ac- 
cording to the determinations on a mixture of the six samples it is 
0"2057, showing a difference, therefore, of only 0'0033 per cent. 
The calculated quantities per acre are, according to the mean 
of the determinations on the six individual samples, 4,062 lbs. per 
acre, and according to the determinations on the mixture 4,132 lbs., 
or a difference of 70 lbs. on a total quantity of about 4,100 lbs. 
Turning to the 1888 results, the difference in the percentage 
of nitrogen in the different samples is rather greater than in 
those of 1879 ; and there is again some, but not an uniform, 
indication of higher percentage with lower weight of soil, and 
lower percentage with higher weight — that is, with probably 
more subsoil included. The mean percentage on the individual 
samples is 0-2423, and on the mixture 0-2405, or a difference of 
only 0'0018 per cent. 
There is also a greater difference in 1888 than in 1879 in 
the calculated amounts of nitrogen per acre from the different 
samples ; but there is very little difference between the average 
result on the five individual samples, and that on the mixture — 
it being 4,618 lbs. according to the analyses and calculations on 
the separate samples, and 4,591 lbs. according to the results on 
the mixed sample — that is, there is a difference of only 27 lbs. 
per acre, in a total of about 4,600 lbs. 
Taking the results as they stand — though, as we shall see 
further on, they will require some modification before being 
finally adopted — the average percentage of nitrogen in the 1879 
samples was 0-2041, but in the 1888 samples 0-2414, or about 
one-fifth higher, after nearly ten years. Or, taking the calcu- 
lated amounts of nitrogen per acre to the depth of 9 inches, they 
were 4,097 lbs. in 1879, and 4,604 lbs. in 1888 ; showing an in- 
crease, therefore, of 507 lbs. in the nearly ten years, or about 
50 lbs. per acre per annum, during so many years of the further 
progress in the formation of the meadow. 
Before referring further to these results, we must endeavour 
to arrive at some estimate of the amount of nitrogen in the 
surface-soil when it was first taken in hand as arable land. As 
