iH Mr Connell's Analysis of several Species of 



potassa ; and, in order to determine whether it also contained 

 soda, the solution, after separation from the small residue of sili- 

 ca, was concentrated by evaporation, mixed with muriate of pla- 

 tinum, and gently evaporated to dryness. It was then treated 

 with alcohol. The alcoholic solution, after being mixed with 

 sulphate of ammonia, was evaporated to dryness ; and the resi- 

 due ignited, and then treated with hot water, to take up what- 

 ever was soluble. This solution, by spontaneous evaporation, 

 yielded a small quantity of prismatic crystals, having all the ap- 

 pearance of sulphate of soda, and efflorescing in the air. From 

 these circumstances, and from the manner in which they had 

 been obtained, there can be no doubt that they were that salt, 

 and the mineral, therefore, contained a little soda ; but as the 

 potassa calculated from the chloride of platinum and potassa left 

 undissolved by the alcohol, and the soda calculated from the 

 sulphate obtained, did not fully make up the total soluble resi- 

 due of the mineral, I have not given the proportions of these 

 alcalies in relation to one another, but have stated the residue as 

 potassa mixed with a little soda, the former alcali constituting by 

 far the greater proportion. 



The quantity of water in the mineral was determined by ignit- 

 ing a portion of it in a platinum crucible. 



The result of the analysis gave, in 100 parts. 



Silica, . 50.14 



Alumina, 17.48 



Lime, ....... 8.47 



Potassa mixed with a little soda, . . . 2.58 



Water, 20.83 



99.50 



Thus this analysis gives nearly the same proportions as those 

 of the chabasites of Ferroe and Gustafsberg already mentioned ; 

 and is also nearly conformable to the composition of the Scottish 

 mineral analysed by Ardfwedson, except in so far as regards those 

 ingredients, which, according to the theory both of the British 

 and of the foreign chemists, contain one atom of oxygen only ; 

 and, with respect to these, that species of replacement men- 

 tioned in the preceding quotation from Berzelius may be held 



