384 F W. Clarke— Theory of the Mioa Group. 



Art. XLY1II. — A Theory of the Mica Group ; by F. W. 



Claeke. 



Ever since its publication in 1878,* in spite of a few dissen- 

 tient voices, Tscheruiak's theory of trie mica group has been 

 generally in vogue. Nevertheless, upon careful inspection, the 

 theory seems open to serious objections. In brief, omitting 

 details for the present, Tschermak regards the micas as made 

 up of four fundamental molecules ; to which the following 

 formulae are assigned : 



1. R' B Al e Si a O,, 



2- Mg i2 Si 6 24 . 



3- H 8 Si 10 O 24 . 



*• F 24 Si 10 O, 



E 7 stands obviously for K, Na, Li, or H ; Mg is equivalent to 

 Fe", and aluminum may be replaced by ferric iron. The first 

 of these formulae represents fairly well the composition of 

 normal muscovite and paragon ite ; the only objection to it 

 being that, as used by Tschermak, it assumes the double atom 

 of quadrivalent aluminum. Since it has been proved by 

 JSilson and Petterson, Combes, Quincke, and others, that alum- 

 inum is truly trivalent, the formula should become R/ 3 Al 3 Si 3 13 ; 

 which, as I have shown in several previous papers, may be re- 

 garded as a substitution derivative of normal aluminum ortho- 

 silicate. 



The second, third, and fourth of Tschermak's formulae 

 represent hypothetical compounds ; the second being a polymer 

 of chrysolite, to which, in nature, some varieties of talc offer 

 the nearest approximation. The third and fourth formulae are 

 analogous to no known substances, and are, moreover, improb- 

 able chemically. " No mica, by itself, resembles in composition 

 either of the three. One formula, then, corresponds to an 

 observed mineral, while the others are purely speculative ; a 

 state of affairs which is not altogether satisfactory. Further- 

 more, the fonr formulae represent as many distinct chemical 

 types, whereas it would seem as if all the true micas should be 

 reducible to one general form, under which their isomorphism 

 would be more readily intelligible. 



There is still another objection to Tschermak's system of 

 formulae, which, if sustained by future investigation, will be 

 fatal. Under his theory the ratio of oxygen to silicon in the 

 micas can never exceed four to one ; while the published 

 analyses of micas often show values far higher, ranging in 



* Ber. Wien. Akad., lxxviii, 5. Zeitschr. Kryst, iii, 122. 



