THE PROPERTIES OF SOILS 199 



increased if all the little clusters of fine particles, which gradually 

 form in clay by the action of the weather, etc., are first of all 

 broken up by kneading the soil in a wet state. The amount of 

 water retained by a soil depends upon the extent of the surface of 

 the particles that get wetted, and for an equal weight of matter 

 the finer the grains are the greater will be their total surface. 



But percolation downwards is not the only motion of the soil 

 water ; it is able also to move upwards in the same way as it will 

 gradually wet the whole of a towel of which only the extreme 

 end is actually dipping into water. To illustrate this action, 

 separate by means of the fine sieve some sand into fine and 

 coarse particles respectively ; take four wide glass tubes eighteen 

 inches or two feet long, tie a little fine muslin over the bottom 

 or plug them with cotton wool and fill them with the coarse 

 and fine sand, the alluvial and the clay soil respectively. 

 Then stand the muslin-covered ends an inch deep in water, 

 and note hour by hour the extent to which water has risen in 

 each, finally plotting the results on a piece of squared paper. 

 The motion is much quicker in the sand than in the clay, but 

 in the coarse sand it extends only for a few inches. In the 

 fine sand the water rises much farther, while in the clay, slow 

 as the motion is, it will continue until the whole contents of the 

 tube are wet. Now these observations can be applied to the 

 study of soils ; at some depth below the surface there is always 

 a layer saturated with water, the level, in fact, at which water 

 stands in the wells sunk thereabouts, and even when this " water 

 table " is at some considerable depth the subsoil a few feet down 

 is none the less highly charged with water. During a drought, 

 as the surface soil loses its moisture, the water will begin to rise 

 from the wet layers below in virtue of the property of capillarity 

 or surface tension that we have just illustrated. When the soil 

 is damp each particle is surrounded by a thin film of water in a 

 state of tension, so that it exerts a pull on other water with 

 which it may be in contact, the pull being greater the thinner 

 and more stretched the film may be. Hence a particle with a 

 thin film will draw water from a thicker film which it touches. 

 As long as the thin films of water coating the soil particles 

 remain continuous and unbroken, water will always creep from 



