140 



with water, increased in binding power and shrinkage, but not in 

 plasticity, whilst colloidal alumina, when similarly treated, increased 

 in plasticity but not in binding j^ower and shrinkage. A mixture of 

 colloidal silica and alumina, made by adding a solution of water- 

 glass to one of alum, increased both the binding power and the 

 plasticity of a clay. Grout rej)eated this experiment, but obtained 

 a negative result. 



On drying the silica gel, the volume decreases to a characteristic 

 point (Van Bemmelen's^' " transition point "), after which it remains 

 constant. On further dehydration, the clear gel becomes turbid and 

 eventually chalk white, but becomes clear again when the water 

 content is reduced to less than one molecule. The addition of water 

 causes little change in volume. At low red heat the gel is completely 

 dehydrated and strong ignition prevents the gel from again taking 

 up water. Any salts present cause the gel to lose its power of taking 

 up water more rapidly. Salts are also favourable to local fusion and 

 destruction of the gel. These characteristics are equally characteristic 

 of plastic clays. 



Colloidal alumina, of which a small and variable proportion occurs 

 in many clays, especially those known as laterites. Colloidal alumina 

 has a very curious and variable effect on the clay. In some cases, 

 it behaves as an acid easily combining with bases to form salts, whilst 

 under other conditions it acts as a base combing with acids. It is 

 soluble in both acids and alkalies forming different compounds in 

 each case. The gel has a particularly low water content, having only 

 about 2J per cent, more than is required to form the hydrate, a fact 

 which is very unusual, as most mineral gels have a very large proportion 

 of free water which can be removed on drying. In most of its 

 physical properties, such as shrinkage, swelling power, &c., it is 

 similar to silica. 



Mixtures of sand and alumina hydrosol, treated so as to coagulate 

 the latter are somewhat plastic, but would not be acceptable to 

 clay workers as a substitute for plastic clay. 



Colloidal Ferric Hydroxide has been isolated in very small quantities 

 from some ferruginous clays, but clearly cannot be an important cause 

 of plasticity in clays which are almost free from iron compounds. 

 Like silica, ferric hydroxide sols increase in viscosity with the concen- 

 tration. The surface tension of .ferric hydroxide sol is the same as that 

 of water and the electric conductivity -was observed by Malfitano 

 to be 200 X 10- «. 



Colloidal ferric hydroxide particles carry a positive charge, i.e., 

 that opposite to that of colloidal silica, so that the two substances 

 mutually precipitate each other. This accounts for some siliceous 

 minerals and clays having a thin coating of iron which also partially 

 fills the interstices. A very peculiar fact with regard to colloidal 

 ferric hydroxide is that it is entirely free from the inky taste which 

 is so characteristic of ordinary iron compounds, and it has no reaction 

 with potassium ferrocyanide (the Prussian blue test which is the most 

 characteristic test for iron), whereas yellowish china clays when 

 treated with ferrocyanide become whiter owing to the Prussian blue 



