THE BKOADBALK WHEAT BOIL8. 
83 
When we examine the figures for the plats actually receiving 
Pblorids, very into nesting differences arc found. Where 4'5 pounds of 
litrogen per acre ifl annually supplied as ammonium salts (half as 
•hlorid). as on plat 6, 50 pounds of chlorin will be included in the 
pressing. On the various plats receiving 86 pounds of nitrogen per 
fore as ammonium salts, the annual quantity of chlorin included is 
\\- pounds, and where 129 pounds of n it roge n: per acre is supplied, as 
>n plat S, ID* pounds of chlorin are included in the dressing. 
If we compare the figures relating to plats A, n. 7, and s, we see 
ihat, though there is practically little difference in the first 9 inches, 
here is, if we take the second and third depths, more chlorin in the 
soil the more there has !>een applied in manure, though the difference 
[snot precisely proportional; and in t he lower (fourth to tenth) depths 
here is, in every case l>ui one, more chlorin found in the soil of plat 
5 than in that of plat 5, more in that of plat 7 than in that of plat 6, 
Knd more in that of plat 8 than In that of plat 7. Taking into account 
the total depth of !M) inches, the chlorin found in plat amounts to 
l'n'.'.'i pounds per acre; that in plat 6 to 195.98 pounds; that in plat 
rtO-241.35 pounds; and that in plat 8 to 318. o7 pounds. Plats (>, 7, and 
3, therefore contain in their ten depths, in round numbers, respect ively, 
['.\ pounds, ss pounds, and 1 »i.*> pounds in excess of what we find in plat 
a numbers which bear on the whole a striking resemblance to the 
|Uantities of chlorin contained in the ammonium salts supplied to each 
plat — namely, 56 pounds, ill' pounds, and 168 pounds, respectively. 
We therefore find, retained in the total ten depths of inches each, 
the chlorin of ten years' rainfall, and in addition approximately that 
>f one year's application of ammonium salts. The chalk is not far 
below these ten dept lis, and t here will doubtless be a larire annual loss 
fcy drainage carrying off bases from the soil combined, a moug other 
forms, as chlorids. It is certainly very remarkable to find that a con- 
stituent whose salts are so soluble as those of chlorin should be 
retained to so great an extent in the soil, or that the amount retained 
should bear Mich obvious relation to the annual sources of supply. 
It would seem that the clay enters into some sort of combination 
with the chlorids from which they are only dislodged by a very free 
•ipplicat ion of water. 
The difficulty of removing chlorids from soil by percolation, except 
when a relatively very Large quantity of water was used, was demon- 
strated in some experiments described in the paper on the rain and 
Lrainage waters at Rothamsted, referred to on p. 49. 
Referring bo tin results recorded for plats 11, 12 13, and 14, where 
he same quant M y of ammonium salts is applied annually with differ- 
m1 mineiai manures, it is seen that the total retention of chlorin, 
low ii to 8 times '.» inches, varies but very little, though considerable 
Variations an found in the first 3 depths. 
Comparing plats 15 and 16, the former of which receives ammonium 
