234 L. V. Pirsson — Mordenite. 



hydrofluoric acids, everything was determined except the silica. 

 The second portion was subjected to a sodium carbonate fusion 

 and everything determined save the alkalies. 



The water was first determined in both portions by ignition. 

 It was given up and the weight became constant at a moderate 

 red heat. Before determining the water in No. 1, it was found 

 that the powdered mineral lost about 3*6 per cent of water by 

 one hour's exposure to a heat of 100° C. The process for 

 determining magnesia and the alkalies was as follows. After 

 separating in No. I, the alumina and ferric oxides by ammonia 

 and the lime by ammonium oxalate, the filtrate was evaporated 

 and ignited gently in a platinum dish until all ammonium salts 

 were driven off. The residue was then dissolved in a little 

 water, and a roughly estimated amount of previously purified 

 barium hydroxide added. By this means all the sulphuric 

 acid and magnesia present were thrown down and the alkalies 

 obtained in the filtrate in a form suitable for conversion into 

 chlorides and for determination, after the excess of barium 

 hydroxide had been removed by ammonium carbonate. The 

 trace of magnesia was then easily separated from the precipitate 

 of barium sulphate by hydrochloric acid, filtered off and deter- 

 mined. 



The analysis was at all points carried on both as a qualitative 

 as well as a quantitative one. It yielded the following results. 



I. II. Mean. Ratio. 



SiO, 







66-40 



66-40 



1-106 



A1„0 3 



11-32 



11-01 



11-17 



•1084 ) 



FeA 



•62 



■52 



•57 



•0036 \ 



CaO 



1-89 



T98 



1-94 



•0346^ 



MgO 



•20 



•14 



•17 



•0042 1 



K„G 



3-58 







3-58 



•0379 [ 



N"a a O 



2-27 







2-27 



•0366 J 



H 2 



13-31 



13-31 



13-31 



•7394 



1-106 9-00 



•112 -91 



•113 -92 



•7394 6-01 



Total, 99-41 



From these ratios it will be seen thart the mineral agrees 

 closely with How's general formula RO, A] 2 3 (Si0 2 ) 9 6H 2 and 

 if the slight amount of magnesia is taken as replacing lime it is 

 evident that the protoxide bases are CaO : Na 2 : K 2 : : 1 : 1 : 1. 

 The composition is then (|K 2 0, iNa 2 0, ^CaO) A1 2 3 (Si0 2 ) 9 

 6H 2 0. In the type of mordenite analyzed by How there was 

 only a slight trace of potash, and his composition showed 

 (-P"a 2 0, f CaO) Al 2 O s (Si0 2 ) 9 6H 2 and in the present mineral 

 one molecule of K 2 replaces one of CaO in How's type. The 

 ratios show a slight deficiency of the bases. For the sake of 

 comparison we give below the theoretical percentages calcu- 



