20 EFFECT OF WATER ON EOCK POWDERS. 



in Table II it was supposed that the absorptions would be deter- 

 mined not only by the valence of the ions but b} 7 the character of the 

 colloids that were present. If clays contain only aluminum silicate 

 and free silica they would be entirely negative and would absorb 

 only basic ions. If, on the other hand, they also contain, as undoubt- 

 edly many clays do, free alumina and iron hydroxid, the equilibrium 

 would be changed and absorptions of both positive and negative ions 

 might take place. The white china clay used in the experiments 

 above described was a fine-grained kaolin which had the following 

 composition: 



Per cent. 



Silica (SiO 2 ) , 56.87 



Alumina (A1 2 O 3 ) - 30.20 



Titanium oxid (Ti0 2 ) 1. 40 



Calcium oxid (CaO) .77 



Magnesium oxid (MgO) Trace. 



Potash (K 2 O) 71 



Soda(Na a O) 17 



Phosphoric acid ( P 2 O 5 ) 24 



Water (H 2 O) 9.82 



Total 100. 18 



Since the absorptions were confined to the positive ions a or bases, 

 this cla} T is acid or negative in character. When a clay of this kind, 

 which contains, as this one does, small portions of the alkaline and 

 alkaline earth bases, such as potash, soda, lime, and magnesia, is 

 treated with dilute neutral solutions of electrolytes which contain 

 none of these elements, such as ammonic chlorid and barium chlorid, 

 not only do exchanges take place, the first-named bases passing to a 

 certain extent into solution, but also the filtrates will usually be found 

 slightly acid in reaction. This is an important point, as will presently 

 appear. If, on the other hand, rocks such as orthoclase, containing 

 an alkaline base (potash), are ground in a ball mill with solutions of 

 ammonic chlorid, the resulting dough invariably smells of free 

 ammonia at the end of the run. This is a sufficient proof in itself 

 that the alkaline bases are set free by the action of water, even if this 

 were not shown to be the case by the phenolphthalein test. Why, then, 

 does not this alkali, which is set free to the extent that it can react 

 upon ammonic chlorid and phenolphthalein, pass freely into solution? 



We are in position now to apply the principles developed in the 

 foregoing pages to the discussion of the action of water on rock pow- 

 ders, for which purpose, to simplify the matter, orthoclase may be 

 taken as a type. There can be no doubt that orthoclase is, at least to 



It is well known that positive colloidal precipitates like ferric hydroxid absorb 

 chlorin from solutions of neutral chlorids. Compare, Chem. Centrbl. 1905, 76, 

 Bd. I, 500. 



