380 The Flocculation of Soils. Ill 



Discussion. • 



The jlucculation of particles other than soil particles. 



The only substances, in the random collection examined, which 

 showed the same anomalous flocculation by calcium hydroxide, as clay, 

 were certain phosphates. Su.spensions of the phosphates of aluminium 

 and of iron, as well as .some natural calcium phosphates, arc flocculated 

 by calcium salts very much better from alkaline than from neutral 

 suspensions. 



The similar behaviour of clay and of these phosphates seems to lend 

 support to the explanation of the abnormal flocculation by calcium 

 hydroxide which has been advanced in the earlier papers. Calcium salts 

 added to a dilute ammoniacal solution of a phosphate, produce a volu- 

 minous gelatinous precipitate very similar to that produced in dilute 

 ammoniacal solutions of sihca. Further, the phosphates of aluminium, 

 iron and calcium are well known to exist quite commonly in the colloidal 

 gel form. The action of calcium hydroxide on the colloidal surfaces of 

 these phosphates is analogous to its action on the colloidal surfaces of 

 silicates and is susceptible to a similar explanation. 



Powdered apatite is crystalhne, without colloidal properties, and it 

 does not show the flocculation anomaly either before or after extraction 

 with cold dilute acid. 



The colloidal condition of phosphates and sihcates is undoubtedly 

 related to their solubility and to their availability to the plant. This 

 aspect of the subject is discussed in a separate paper. 



The respective rdles of the core and of the colloidal surface. 



Assuming the truth of the earlier deduction that calcium hydroxide 

 has a dual action on .soil particles (a deflocculating action on the cores of 

 the particles and a precipitating action on the emulsoid surface) it is clear 

 that the deflocculating action may dominate the precipitating action in 

 at least two circumstances: 



(i) When the amount of emulsoid matter is small relatively to the 

 core. 



(ii) When the amount of calcium hydroxide is insufficient to cause 

 a maximum precipitation with the colloidal matter. 



The first of these circumstances has already been demonstrated and 

 discussed. The second has now been demonstrated experimentally, for it 

 has been shown that when the concentration of the flocculant is very low 

 clay suspensions are flocculated less readily by calcium hydroxide than 



