INFLUENCE OF MOVEMENTS OF WATER 199 



Should the soil in question be rich in porous particles, in 

 humus for instance, the action of the descending water column 

 will be less perfect than we have supposed, the salt solution 

 in some of the pores remaining undisturbed for a sufficient time 

 to affect the result. The first discharge from the soil under 

 these circumstances will consist, as before, of the unaltered 

 soil solution ; but the whole of the soil solution will not be 

 discharged unaltered, the latter portion appearing in a diluted 

 form. 



Schloesing (Chimie agricole, 127) has made considerable 

 use of the method of displacement in order to procure for 

 examination samples of the solutions naturally present in 

 the soil. Working with non-porous sand, he was able to 

 obtain 85 per cent, of the solution which it contained in an 

 unaltered form by displacement with water. With a soil 

 of very varied physical constituents, only about 20 per cent, 

 of the original solution was collected unaltered, the remainder 

 coming through more or less diluted. In Schloesing's experi- 

 ments the water was sprinkled on the surface of the soil, and 

 percolation took place by gravitation only. 



More striking results are obtained when a uniform powdered 

 soil is employed, and the water is made to descend more 

 rapidly by means of an air pump. The following experiments 

 were made by myself in the Rothamsted Laboratory (J. Roy. 

 Agri. Soc. 1881, 329). The percolator used consisted of a half- 

 gallon bottle from which the bottom had been removed ; this 

 was fixed mouth downwards, thus forming a large cylindrical 

 funnel. In this funnel was placed a disk of fine wire-gauze, 

 covered by a rather larger disk of filter paper, and on this 

 7 Ib. of finely powdered, air-dried loam ; the soil was well 

 shaken in, so as to lie as compactly as possible. The column 



