METHODS OF MECHANICAL ANALYSIS n 



in each separated portion of the soil can be ascertained, if 

 this is thought desirable. It is of course equally possible 

 to determine the amount of vegetable residue and humic 

 matter in each of the separated parts. 



The separation of the finest portion of the soil into two 

 groups is also conveniently carried out by using the cylinders 

 devised by Kiihn and Knop (Wiley, Agricultural Analysis, i. 

 189). The soil having had its coarser ingredients separated 

 by sieves, and having been well boiled, is introduced into a 

 tall cylinder, which is then filled with water and well shaken. 

 After a certain number of minutes, a side tube is opened and 

 the turbid water allowed to flow out, and the operation is 

 repeated till the water runs clear. 



The methods we have just mentioned do not attempt to 

 separate all the particles of a soil into groups of definite size 

 or character. The two American methods we shall next 

 describe yield far more complete results. 



In Osborne's method (Rep. Exp. Station, Connecticut, 1886, 

 141 ; 1887, 144; 1888, 154. Wiley, op. cit., 196) the separa- 

 tion of particles of different sizes is effected by a graduated 

 subsidence, as in the methods already noticed ; but the whole 

 operation is controlled by a microscopic measurement of the 

 largest and smallest grains in each sediment, so that the 

 portions finally weighed are known to consist of particles 

 the diameters of which fall within the limits laid down in 

 the scheme. It must be borne in mind however, that any 

 group of particles obtained by subsidence will not be entirely 

 of the same size in cases where the soil particles consist of 

 substances having different specific gravities. This being 

 inevitable, the character of the group is fixed in every case 

 by the diameter of the quartz particles only. The same 



