Washington — Constitution of Some Salic Silicates. 561 



and Bradley* justly criticise this interpretation as being "open 

 to the serious objection that a chemical compound, so far as 

 we. know, does not vary in type. Isomorphons replacement, 

 for instance, varies the composition, but the type of compound 

 remains the same." This remark applies equally well to the 

 usually accepted formulas for the end members of the feld- 

 spars and scapolites, discussed above. Foote and Bradley con- 

 sider that the high silica of nephelite is due to the presence of 

 silica in solid solution in NaAlSi0 4 . Later Schallerf suggests 

 that the composition is to be explained as an isomorphous mix- 

 ture of the crystallographically hexagonal forms of the com- 

 pounds NaAlSi0 4 , KAlSi0 4 , and a hypothetical hexagonal and 

 isomorphous NaAlSi 3 8 . This explanation rests primarily on 

 the assumption that (Si0 4 ) and (Si s 8 ) are isomorphously 

 replaceable. This view and the practical equivalence in func- 

 tion ojf these two acid radicals has been especially insisted on 

 by F. W. Clarke, but is open to the general objection noted 

 above of non-conformity with the modern theory of isomor- 

 phism. It is, indeed, difficult to conceive of the mutual iso- 

 morphous replaceability of a trisilicic and an orthosilicic acid 

 radical, and, as will be seen, such a conception is not necessary 

 in the case of the mineral groups under discussion. In a 

 paper just published Bowen^: comes to the conclusion, based 

 largely on the preparation of an artificial nephelite, that the 

 explanation suggested by Schaller is correct. 



The suggested hypothesis. — Streng§ seems to have been the 

 first to appreciate the non-conformity of the then accepted 

 formulas of the end members of the feldspars with the theory 

 of isomorphism as enunciated by Mitscherlich, and in order to 

 render them of the same type suggested the doubled formulas 

 Na 2 Si 2 .Al 2 Si 4 16 for albite and Ca 2 Al 2 .A] 2 Si 4 16 for anorthite. 

 These were adopted and later abandoned by Tschermak. In 

 the different editions of his Tabellarische Uebersicht, Groth 

 suggests various structural formulas for the feldspars, and 

 wavers between two types of constitutional formula, namely : 

 NaAlSiSi 2 8 and CaAlAlSi„O e (1882, p. 110), and Si 2 Si0 8 AlNa 

 and Si 2 A10 8 AlCa (1889, p. 137, 1893, p. 156). These, espe- 

 cially the first pair, but somewhat differently written, are 

 essentially those upon which he finally!! settles, namely, 

 (NaSi)AlSi 2 8 for albite (and a corresponding one for ortho- 

 clase), and (CaAl)AlSi 2 8 for anorthite. These formulas are 

 of identical chemical type, the quinquivalent integrals (NaSi) 



* Foote and Bradley, this Journal, vol. xxxi, p. 30, 1911. 

 f W. T. Schaller, Jour. Wash. Acad. Sci., vol. i, p. 109, 1911. 

 JN. L. Bowen, this Journal, vol. xxxiii, p. 49, 1912. 



§Streng, Neues Jahrb., 1805, pp. 411, 513. Cited by Hintze, Mineralogie, 

 vol. ii, p. 1431, where a resume of the early discussion will be found. 

 I P. Groth, Introd. Chem. Cryst., 1906, p. 86. 



Am. Jour. Sci. — Fourth Series, Vol. XXXIV, No. 204. — December, 1912. 

 37 



