560 Washington — Constitution of Some Salic Silicates. 



the replacement being between elements or atomic groups of 

 like valency, either residual or total, and the physical proper- 

 ties being continuous functions of the composition. 



Based on these criteria the sodi-calcic feldspars form a per- 

 fectly isomorphous series so far as their crystallographic and 

 physical properties, both scalar and vectorial, are concerned. 

 This is abundantly shown by many researches, especially those 

 of Day and his colleagues,* and Grothf also . points out that 

 they possess almost identical equivalent volumes. Similarly 

 the orthoclase-celsian series and the scapolites, chabazites, and 

 other zeolite groups, form crystallographically and physically 

 typical isomorphous series of mineral salts. 



But from the chemical point of view the usually accepted 

 formulas of the end members, of which we have every reason 

 to believe that these several series are composed, are not iden- 

 tical in type and therefore do not conform with the definitions 

 and fundamental conceptions of isomorphism as generally 

 understood. Thus the formulas of albite, NaAl 2 Si 3 O e , and 

 anorthite, CaAl„Si 2 8 , are not of the same type, one being that 

 of a trisilicate and the other of an orthosilrcate. The same is 

 true of orthoclase and celsian. In the scapolites the formulas 

 of the end members as interpreted by Tschermak are open to 

 the same objection. Marialite, Na 4 Al 3 Si 9 2l Cl, and meionite, 

 Ca 4 Al 6 Si 6 Q5 , are not identical in chemical type and are there- 

 fore not in accord with the accepted criteria of isomorphous 

 substances.:}: The same objection may be brought against 

 the formulas suggested by Groth, Na^AlCljAl^SigO^ and 

 Ca 4 (A10) Al s (Si 2 A10 8 ) 3 . Similar non-conformity with the 

 theory of isomorphism is found in the accepted formulas of 

 the end members of the phillipsite and chabazite groups 

 among the zeolites. 



The complex and somewhat variable composition of nephe- 

 lite, with its content of potash and excess of silica, is appar- 

 ently inconsistent with its evident crystallographic isomorphism 

 with kaliophilite, eucryptite, and artificial, purely sodic nephe- 

 lite. The composition of nephelite lias been variously 

 explained. Following the suggestions of Rammelsberg and 

 Doelter it is usually assumed that a leucitic molecule is present. 

 More recently Morozewicz§ suggests that the nephelites are a 

 series of compounds of the general type K„lSTa n Al n+:J Si D+3 O 4n+10 , 

 with a less common basic series. The best analyses of nephe- 

 lite can be interpreted in terms of these molecules, but Foote 



*Day, Allen, etc., Carnegie Publication, No. 31, 1905. 



+ Groth, Introd. Chem. Cryst., New York, 1906, p. 86. 



X It is noteworthy that the same criticism applies to the formulas sug- 

 gested by Tschermak for the end members of the pyroxenes, amphiboles, 

 micas, and chlorites. 



§ Morozewicz, Bull. Acad. Sci. Cracovie, 1907, p. 958. 



