116 ANNALS NEW YORK ACADEMY OF SCIENCES 



eSiOj, to anorthite, CaOAl2032Si02, form, therefore, a single chemical 

 phase. While natural plagioclases are, in the great majority of instances, 

 combinations of albite and anorthite, there are reasons for thinking that 

 soda anorthite (jSTagOALOaSSiOo) and lime albite (CaOAlsOgGSiOa) are 

 also possible compounds, Doelter showed a number of years ago that 

 fusion and slow recrystallization of some of the zeolites resulted in the 

 formation of the compound CaOALOgGSiOo, and H. S. Washington and 

 F. E. Wright^" have lately investigated the composition of a feldspar from 

 Linosa, whose formula from analysis corresponds to (^ Ca, f Na) ALSig- 

 Oio- They consider the most probable explanation of composition, on 

 chemical and crystallographic grounds, to be the presence in solid solu- 

 tion of Ab, Or, An and a new compound, N'a20Al2032Si02, for which 

 they propose the name carnegieite. They point out the resemblance of 

 the mixed crystals which compose the feldspar to an unhydrated natrolite 

 or mesolite. The matter has an important bearing upon the relations of 

 the zeolites to the feldspars. 



Quartz (SiOs), prehnite (H2CaoAl2Si30i2 or 2CaOAl2033Si02H20) 

 and pectoUte (HNaCag (8103)3 or N"a204Ca06Si02H20) form each a 

 phase of invariable composition. They are composed of the same oxides 

 as the other members of the group which we are considering, but they do 

 not present the same analogies in chemical structure. 



The zeolites proper are all feldspathoid compounds whose resemblance 

 in composition to the plagioclases has often been commented upon. The 

 ratio of lime plus soda to alumina is 1 : 1, but the amounts of silica and 

 water vary. They and the feldspars might all be embraced under a 

 general formula EOALOsmSiOgnHoO. With certain of the species, the 

 composition appears to be fixed, as inlaumontite (CaOAl2034Si024E[20), 

 but the majority of species form groups whose composition varies within 

 rather wide limits. A typical example of this is the chabazite group, 

 whose variability appears only explicable on the assumption that four 

 distinct feldspar-like molecules enter independently (CaOAL032Si02- 

 4H2O), (Iia20Al2032Si024H20), (CaOAl2036Si028H20) and 1^x6- 

 Al2036Si028H20). The resemblance in composition which the various 

 molecules bear to albite and anorthite and the assumed calcium-albite 

 and sodium-anorthite is evident, and, like the ordinary plagioclase mole- 

 cules, they appear to possess unlimited mutual solubility and form but a 

 single phase. The structure of stilbite is similar, but less complex. The 

 essential molecule appears to be CaOALOgGSiOoGHoO, with which is 

 united in small but variable proportions the corresponding soda molecule. 



' Am. Jour. Sci., 4th ser., vol. 29, pp. 52-70, 1909. 



