METHOD OF SEPARATING THE CLAY AND SAND IN CLAY SOILS. 101 



admixture of clay. On the other hand, by the water pro- 

 cess the residual "sand " was almost entirely made up of 

 clay floccules of the dimensions of medium sand grains, 

 each floccule or aggregate appearing to be composed of 

 minute particles of decayed vegetable matter to which 

 adhered clay and particles of sand. The heavier sand grains 

 were observed to settle down rapidly after the current of 

 water was stopped, but the major portion of the sand was 

 distributed throughout the clay etc. which deposited more 

 slowly. In order to compare the disintegrating power on 

 the clay floccules, of the zinc chloride reagent, and the 2°/» 

 alkali solution recommended by Schone 1 for soils rich in 

 humus, 30 grams of the same soil were boiled in 250 cc. of 

 a two per cent, solution of caustic soda for 1 hour, the 

 gravel etc., having been previously removed as in the other 

 cases. 



Great care was necessary to keep stirring before coming 

 to the boil, as the soil becomes very flocculent and settles 

 rapidly. The elutriation took 8 hours, the dried residue 

 weighing 6*4 gram equivalent to 21*3°/° of sand in the soil. 

 The appearance of the "sand" presented the same defects 

 as those observed in the water process, though in a lesser 

 degree. Apparently the effect of the alkaline hydrate is 

 beneficial to a certain extent, dissolving the humus which 

 has a binding effect on the clay particles, but exerting no 

 solvent. action on the vegetable fibre itself (cellulose). The 

 superiority is apparent therefore of a reagent that will 

 eliminate both these causes of adhesion of the particles. 



In order to test the value of strong nitric acid in this 

 direction, 30 grams of the fine soil were boiled in the strong 

 acid for one hour. The reaction in this case is very violent 

 and there is great difficulty in preventing the whole from 

 frothing out of the vessel, great care being required, 



1 Wiley Agr. Analysis, Vol. i., page 219. 



