G. F. Becker — Silicic Acids. 155 



agine diopside (for example) as a calcium salt of A combined 

 with a magnesium salt of B, and tremolite as a magnesium salt 

 of A in combination with a calcium salt of B. Uralitization 

 would then be represented as an interchange of bases between 

 the two acids. The modulation from one crystallographic sys- 

 tem to another is no doubt due to substitution, however these 

 metasilicates may be regarded, but the dimorphism which ex- 

 tends throughout the two series yielding pyroxenes and amphi- 

 boles of similar composition, shows that the basic molecules 

 are united with the acid groups in two distinct ways and, in 

 the non-aluminous members of the series, it is difficult to see 

 how this can be unless the acid is complex. 



Were the metasilicates and particularly the pyroxene-am phi- 

 bole group to be considered as ortho-polysilicates they would 

 present a close parallelism to the composite feldspars. Nearly 

 all mineralogists now accept the theory of isomorphous mix- 

 tures of anorthite and albite. The former is of course an 

 orthosilicate and the latter a polysilicate. These isomorphous 

 mixtures would seem to represent in reality some slight chemi- 

 co physical reaction between the components for, if no energy 

 were depotentialized in the formation of the mixtures, these 

 would have no tendency to form, and if energy is dissipated 

 by the union of albite and anorthite, some action (perhaps com- 

 parable to solution) must occur. That these isomorphous mix- 

 tures should be accompanied in nature by double salts of the 

 same acids, but in part containing different bases, seems very 

 natural. The hypothesis that the metasilicates are double salts 

 evidently does not preclude the theory that they too are some- 

 times isomorphous mixtures. 



The studies of Messrs. Brush and E. S. Dana on the spod- 

 umene of Branchville* show that when this pyroxene under- 

 goes decomposition, a part of the lithium is replaced by so- 

 dium, and that the mineral breaks up into eucryptite and albite. 

 This decomposition can be neatly exhibited in accordance with 

 the hypothesis here suggested by means of structural formulas 

 similar to those which Prof. F. W. Clarke has so effectively 

 employed. Spodumenet and its decomposition-products may 

 then be represented by 



/Si 2 6 = Li s /Si 3 8 = Na 3 /Si0 4 =Li 3 



Al— Si 2 O e = Al Al-Si s 8 ^Al Al-Si0 4 = Al 



NSi a 6 = Al \Si 8 8 = Al \Si0 4 = Al 



Spodumene. Albite. Eucryptite. 



so that two molecules of spodumene, as here written, are re- 

 solved into one molecule each of the other minerals when one 



* System of Mineralogy. App. III. f Amer. Chem. Journ., vol. x, No. 2. 



