Mr. Northcote on the Constitution of Thermoj)hyllite. 265 



These results correspond closely with the following formula : — 



-37MgO . Si03 

 4NaO . Si03 



49MgO.HO+<; j,gQ- 



L 4H0 . AFO" 



2H0 "^•'O' 



which has the following per-centage composition : — 



Soda 2-70 



Magnesia 37-40 



Alumina 5'58 



Protoxide of iron . . . 1*56 



Water 10-76 



Silica 42-00 



100-00 



Tliis formula may appeal cumbrous ; but if it be allowed that 

 a true substitution of magnesia can take place by soda, which 

 mineralogists frequently assume; and by water, which is certainly 

 not more improbable ; and that silicic acid (SiO^) may be re- 

 placed by the triatomic aluminic acid which exists in Gibbsite 

 (SHO.AFO^), and by the monatomic aluminic acid which is 

 found in diaspore (HO . Al- 0'^) ; then it will be found that this 

 mineral thermophyllite is essentially 



MgO.HO + MgO.SiO^; 



the silicate of magnesia being partly replaced, at least in this 



specimen, by the silicates of soda and protoxide of iron, by the 



monobasic aluminate of water, HO . AP 0^ (diaspore), and by 



FeO 1 

 the tribasic aluminate of iron and water, oHO r-^^^O^ ^ body 



which, although not known, I believe, in the separate state, has 

 its existence rendered highly probable by the occurrence of the 



analogous aluminate of potash, 2 jiQ ^APO^, which has been 



obtained artificially in crystals. The simple fact of any given 

 salt not having been found as a mineral is not a sufficient war- 

 rant for the assumption that it does not so exist, especially 

 when it is in striking analogy with other known compounds, of 

 the existence of wliich no question can be raised. 



It would be interesting to examine other specimens of this 

 mineral, to ascertain whether the typical formula 



MgO.HO + MgO.Si03 



ever exists, or whether other replacements exist which render 

 that view of its composition admissible. 



