58o SCIENCE PROGRESS 



tube of soil could then be used as a specification of the soil, 



so that different soils could be classified simply by comparing 



the numerical values of the particle radii of the equivalent 



" ideal " soils. This treatment had the obvious advantage 



that a single figure could be regarded as specifying a natural 



soil, but it also possessed the disadvantage that the sand-grains 



constituting the " ideal " soil with which the comparison was 



made were not spherical, nor were they all of the same diameter. 



Green and Ampt, in Australia, were fortunate in finding 



a material which did not suffer from these defects. This is 



the so-called " Glistening dew " of the Christmas-card artist, 



which is composed of almost perfectly spherical glass grains 



of diameter -25 mm. upwards, and is an excellent material for 



an " ideal " soil. Green and Ampt measured the permeability 



(i.e. the volume of fluid passing per second through a soil 



column of i sq. cm. cross sectional area, and i cm. in length, 



under i cm. head of pressure) of the " ideal " soil both for air and 



water, and made similar determinations for actual soils. Now, 



it can be shown that the actual permeability (P) varies inversely 



as the viscosity {/x) of the fluid used, provided there is no 



action of the fluid on the material ; so that for air (a) and 



u P 

 water (w) the value of the ratio -^ should be unity. With the 



glass bead " ideal " soils, this relation was obeyed, but for the 

 actual soils the value of the ratio ranged from 2 in a sandy 

 soil up to 14 for a clay soil. This is due to the imbibition 

 of water by the colloidal material, which increases in volume 

 and thus decreases the size of the capillary spaces. Low values 

 are therefore obtained for Pw, which are reflected in the high 



value of the ratio ^^^^. These experiments of Green and Ampt 



emphasise the important difference between the constitution 

 of actual soil and a mass of sand-grains. 



We have mentioned above that the tilth of a clay soil 

 may be damaged by cultivation when wet. The structure of 

 the compound particles is broken down, so that the individual 

 grains are forced closer together and the whole soil becomes 

 sticky and plastic ; if it subsequently dries the soil contracts 

 considerably into hard coherent masses, separated by cracks, 

 which may be several inches wide and of considerable depth. 

 The plasticity of wet soil and the cohesion of dry soil, therefore, 

 are properties which must be studied in their relation to the 

 formation of compound particles. 



There is great scope in this direction for further work, 

 especially in view of the relations of these phenomena to the 

 various methods of cultivation used in modern farming. For 

 convenience of discussion, consideration of cultivation imple- 



