292 CAUSES OF FERTILITY AND STERILITY [chap. 



weight of an acre of soil an inch thick, it is obvious that 

 the difference in water content so induced would not be 

 sensible. 



Clay soils, in fact, which have been treated with 

 nitrate of soda, do not show any excess of water ; but 

 they are very much deflocculated, as may be ascertained 

 by comparing the appearance after standing of a jar of 

 distilled water rendered turbid by shaking up in it a 

 gram of the soil, and a second jar in which the water 

 has been shaken with a gram of the same soil in its 

 normal condition. But nitrate of soda itself possesses 

 flocculating powers even when concentrated, hence the 

 observed deflocculation can not be due to the direct action 

 of the fertiliser upon the clay. However, it has been 

 found that when plants feed upon a nutrient solution 

 containing nitrate of soda, an excess of the nitric acid 

 is withdrawn by the plant, and part of the soda is left 

 in the medium combined with the carbon dioxide 

 secreted by the plant. The existence of this soluble 

 alkali after the growth of the plant can be verified by 

 experiments with water cultures, it can also be extracted 

 from the soil of the Rothamsted plots which have for 

 many years been manured with nitrate of soda. Small 

 as the amount may seem to be, it is quite sufficient to 

 account for the deflocculation of the clay and the 

 defective tilth observed on heavy land after nitrate of 

 soda has been used. 



The bad repute of nitrate of soda as exhausting or 

 scourging the land, is less due to any sensible diminution 

 in the stock of plant food in the soil that follows its use, 

 than to the deflocculation it sometimes induces, and the 

 consequent deterioration of the texture of the soil. 

 As a remedy lime is not effective, since it is an alkali 

 itself; instead, the nitrate of soda should be used in 

 conjunction with acid flocculating manures like super- 



