jr/te History of a Field newly laid down to Permanent Grass, 13 
six 1879 samples, and in the lower division for each of the five 
1888 samples, the calculated amounts per acre of the total soil 
as sampled, of the stones, the roots, the water, and, finally, of 
the fine soil free of water. 
It will be seen that, according to the calculations, the average 
weight of total soil per acre, to the depth of 9 inches, was about 
330,000 lbs., or about one-eleventh less in September 1888 
than in January 1879. An examination of the amounts of 
water shows that about two-thirds of the difference was due to 
the less amount of moisture in the soil of the 1888 samples, 
taken in September, and after the removal of a second crop of 
hay, than in those taken in January 1879, six months after the 
removal of the crop. The remaining difference, amounting to 
rather more than 100,000 lbs., is made up by rather over 3,000 
lbs. less stones, rather over 1,000 lbs. less roots, and very nearly 
100,000 lbs. less dry soil ; corresponding, in fact, to about 
5 per cent., or one-twentieth less dry soil in 1888 than in 1879. 
It may be observed that the difference in the amount of 
water per acre, to the depth of 9 inches, at the two periods, 
corresponds to a difference of almost exactly 1 inch less rain 
retained in the first 9 inches of depth in September 1888 than 
in January 1879 ; and the less amount of soil removed at the 
later date is probably, in great part, due to less consolidation 
of the drier soil. The average amounts of water represent about 
20-8 per cent, of the soil as sampled in 1879, but only lG-1 per 
cent, in 1888. 
It is seen that, on the average, the stones amounted in 1879 
to nearly one-fourth, but in 1888 to more than one-fourth of the 
total soil as sampled ; and, in each case, the stones amounted to 
not far short of half as much as the fine diy soil, which is a 
larger proportion than has been found in any, and a much 
larger proportion than in most, of the surface-soils, of the Roth- 
amsted experimental arable fields. The amounts varied, how- 
ever, considerably according to the different samples, and more in 
those taken in 1888 than in those of 1879. 
The amount of roots also varied considerably in the different 
samples, but more in 1879 than in 1888; the quantity cor- 
responding in one case in 1879 to 8,875 lbs. per acre, and in 
another to 16,008 lbs. ; whilst in 1888 the greatest difference 
was from 7,623 lbs. in one case, to 12,523 lbs. in another. On 
the average there was about one-eleventh less roots in 1888 
than in 1879 ; whilst we should expect an increase with the 
progress in the formation of the meadow. Part of the difference 
is accounted for by the less amount of total soil taken to the 
depth of 9 inches in the drier state of the soil in 1888 ; and it 
