202 V-'Cr ) 'rTJAX . L UJ! ICULTUR /;. 



drains are kept at far too high a level. If drainage had 

 to be pumped the cost of surface washing would be pro- 

 hibitive. 



When the level of water in a drain can be kept down 

 75 centimetres below the surface of the land, then by the 

 filtration method a 7 5 -centimetres layer of soil can be 

 sweetened, the drains carry little water and that water 

 holds much salt in solution, whilst little irrigation water 

 has been used. If the same land were washed by the 

 surface method, the drains would stand nearly full and 

 not above 25 centimetres of the surface soil would be 

 sweetened. 



It is true that by merely passing large quantities of 

 water over the surface of the land the soil will become 

 sweet, even without any drains being made. There are 

 examples of that wherever a .canal or comparatively sweet 

 drain flows over the marginal land of any of the lakes. 

 Around all such escapes, reeds and coarse grasses spring- 

 up. The process however is exceedingly slow. The 

 action which here comes into play is that of the diffusion 

 of liquids. The salt water of the lower stratum gradually 

 works to the sweeter water above, mingles with it and is 

 carried off, and so salt is gradually eliminated. The salt 

 water table not being permanently lowered and kept down 

 by drains, as soon as the supply of fresh water ceases, 

 capillarity quickly brings up salt and the lands revert. 

 It is only where there is constant escape that reeds flourish. 

 Where the escape ce"ases reeds do not grow and it is not 

 for want of water, for some kinds will last a summer 

 without irrigation. It is because the salt returns whoiv 

 there are no drains. 



