2<So SCIENCE PROGRESS. 



aluminium. Kaolin then has the composition (HO) 4 AL 

 Si 2 5 and Orthoclase is (Si 2 5 )(K Si 3 ) 2 Al 2 Si 2 5 , and both 

 may be regarded as salts of dimetasilicic acid H 2 Si 2 O r 



Clarke, on the other hand, is concerned to prove that 

 the natural silicates are all substitution derivatives of normal 

 salts (mostly of orthosilicates) ; he therefore regards Ortho- 

 clase as having the composition K 3 Al 3 (Si 3 8 ) 3 , i.e., as a normal 

 salt of the trisilicic acid H 4 Si 3 O s . This acid he considers as 

 equivalent to orthosilicic acid H 4 Si 4 , and the acid radicles 

 as mutually replaceable. Such a supposition is conveniently 

 invoked to account for the undoubted fact (in accordance 

 with Tschermak's theory of the Felspars), that Albite 

 Na Al Si 3 O s and Anorthite Ca Al 2 (Si 4 ) 2 are isomorphous 

 and mix in all proportions. The relationship between 

 Orthoclase and Kaolin is then shown in Clarke's interpreta- 

 tion by attributing to the latter the formula H 3 Al 2 (OH) 

 (Si 4 ) 2 . Kaolin is in some respects closely allied to the 

 Micas, and Clarke compares its constitution with that of 

 the Lime-iMica, Margarite, (Ca H)(A1 0) 3 A1(0 H)(Si0 4 ) 2 . 



Groth, for the sake of simplicity, regards Kaolin as a 

 dimetasilicate in which the two hydrogen atoms of the acid 

 are replaced by two univalent radicles A1(0 H) 2 , and refers 

 Serpentine, Talc, Picrosmine and Pyrophyllite to the same 

 acid as allied minerals. 



(2) Talc H 2 Mg 3 Si 4 I2 is a very common result of the 

 decomposition of magnesian silicates. 



The alteration of Enstatite Mg Si 3 to Talc suggests 

 to Scharizer that Enstatite has the composition Mg 4 (Si 3 ) 4 , 

 and Talc H 2 Mg 3 (Si 3 ) 4 , so that the change is simply a re- 

 placement of Mg by H 2 in a normal metasilicate. Clarke, 

 reasoning from the fact that upon ignition one-fourth of 

 the silica of Talc is liberated, prefers to regard this mineral 

 as Mg(Si0 4 )Mg 2 (Si 3 8 )H 2 or H 2 (Si 4 )Mg 2 (Si 3 8 )Mg, i.e., 

 as a pseudometasilicate which is to be taken as a com- 

 pound orthosilicate and trisilicate. Enstatite he would 

 therefore regard as a similar compound orthosilicate 

 and trisilicate Mg 4 (Si 4 )(Si 3 8 ). Groth had previously 

 suggested that Talc might be a basic dimetasilicate 

 Mg(Mg OH) 2 (Si 2 5 ) 2 , while Tschermak, in order to show 



