134 
INVESTIGATIONS ON ROTH A MSTED SOILS. 
It will be remembered that, of the ammonium sails used a1 Rotham- 
sted, one-half of each application consists of ammonium chloric!; and 
that, on the plats so dressed, the quantity of chlorin found, in excess 
of that on the unmanured plats, was in 1893 very fairty proportional 
to the quantity added in the ammonium salts. 
It will also be remembered that, when the 1893 chlorin results were 
discussed, the curious fact was pointed out that when the soils and 
subsoils were examined to as far down as ten successive depths of 
9 inches each it was found that each plat contained a quantity of 
chlorin equivalent to that applied in one year's manurial dressing, 
pins the average chlorin of ten years' rainfall — an average year's 
rainfall at Rothamsted containing 14.75 pounds of chlorin per acre. 
Notwithstanding the constant percolation of rainfall, each 9 inches of 
subsoil is still found to contain, even when no chlorin is applied in 
manure, about 15 pounds of chlorin per acre, according to the results 
obtained in the analyses of the 1893 samples. 
In comparing the 1881 and 1893 results, taking in the aggregate 
the three depths down to 27 inches, we find very little difference on 
the unmanured plats, 3 and 4. Plat 5 in 1881, as in 1893, was found to 
contain more chlorin than the unmanured plats, although not more 
than a few pounds per acre per annum are applied to it as impurities 
in the mineral fertilizers used. 
On plats 6, 7, and 8 we find in 1881, as in 1893, a successive increase 
in chlorin as the quantity of ammonium salts grows successively 
larger; and the other results are for the most part fairly consistent. 
It is noticeable, however, throughout the ammonium-dressed plats, 
thai notwithstanding the very much greater rainfall in the summer 
and autumn of 1881, the quantity of chlorids in these plats was found 
in October to be uniformly greater in 1881 than in 1893. As we have 
already seen, it is highly improbable that any tangible quantity of 
actual ammonium salts was left in the soil in the autumn of 1881, and 
the chlorin added in the form of ammonium salts must by that time 
have existed in mineral combination with calcium, magnesium, or other 
bases in 1 lie soil. The quantity of chlorin present in 1881 as com- 
pared with that found in 1893 was not only greater when the 27 inches 
were considered, bul was even much greater in the surface soil than 
in the corresponding soils of 1893, as well as being, as a rule, much 
higher in t he second ami t hird depl hs. In fact, if we deduct the yield 
of the unmanured plats, we find thai as a rule the quantity of what 
may be called manurial chlorin found in the firsl 27 inches of soil and 
subsoil in 1881 was two or three times as great as in 1893. It would 
seem that, in 1 ssi chlorids which must have been initially washed 
farther down had, by a process of upward diffusion, readjusted them- 
selves in the soil water under the influence of the evaporation pro] 
duced by early aut umnal heat. 
