INFLUENCE OF MOVEMENTS OF WATER 203 



diffused downwards to a considerable extent during the ten 

 days that had elapsed since its application ; for had the 

 chloride remained entirely at the surface it would have 

 required 850 grams of water to carry it through the soil, 

 this being the quantity of water required in the previous 

 experiment to saturate this mass of soil, and consequently 

 the amount necessary to displace the water contained in the 

 soil when already saturated. 



Fresh proof of the activity of diffusion is furnished by the 

 composition of the successive discharges of drainage water. 

 The drainage is not at its maximum strength at first, as in the 

 previous experiment. The chlorides when they first appear 

 are small in quantity, and the strength of the discharge rises 

 rapidly during four days, and then diminishes in an equally 

 gradual manner. A glance at the proportion of chlorine per 

 million of water shows that a band of salt solution was 

 passing downwards through the soil during the experiment, 

 and that the salt was during the whole time actively diffusing 

 from the upper and lower surfaces of this band into the soil 

 saturated with water through which it was moving. 



The slower is the passage of water through the soil, the 

 larger is the quantity of water required to remove the soluble 

 salts. In the first experiment, all the chlorides were removed 

 from the soil in four hours, by the use of 1,000 grams of 

 water. In the second experiment, lasting eighteen days, 1,320 

 grams of water were required to accomplish the same work. 

 The effects produced by diffusion become very considerable 

 whenever time is allowed for their manifestation, and these 

 effects are in the main conservative, and tend to counteract 

 the removal of salts from the soil by rain water. Very heavy 

 or continuous rains are very injurious to fertility, the surface 



