III.] COHESION DUE TO SURFACE TENSION S3 



point if the brush be dipped in water and withdrawn. 

 Or again, a flat sandy beach from which a smooth sea 

 is receding will often show above tide-mark one stretch 

 of sand quite dry and loose, in which the feet sink 

 deeply, and another very soft stretch immediately left by 

 the tide where the sand grains are completely sur- 

 rounded by water. Between the two is a stretch of 

 sand of the same character, but firm to walk upon; 

 this is partly wet, and there is enough water to form a 

 film round the grains and hold them in position with a 

 certain amount of force. That this sand is really just 

 as loosely arranged as the softer tracts that are either 

 wetter or drier, may be seen by the fact that it will 

 easily pack more closely together under repeated gentle 

 pressure with the foot The shrinkage of soils, especially 

 of clays, as they dry, may be attributed to the surface 

 tension of the films surrounding the groups of soil 

 particles; as the water content is lessened the films 

 exert more force in their effort to contract, and drag 

 some of the particles closer together, especially the very 

 small particles whose weight is trivial compared to the 

 forces exerted by the film. Clay shrinks more than 

 other soils because of the greater number of particles, 

 their small size, and the higher proportion of pore space 

 into which motion can take place. The tenacity of wet 

 clay is due to the number of water films that have to be 

 ruptured, the vastly greater cohesion of dry clay 

 probably to the fact that many of the particles have 

 been dragged within the range of one another's 

 molecular forces. There is a stage in the drying of 

 clay when it will fall to pieces when worked ; probably 

 this represents the stage analogous to the partly wet 

 sand, when cohesion is due to the surface films. The 

 clay is neither so wet that the particles just slip over 

 one another when pressure is applied — the pasty 



