J 119 



cuie the material for a chemical analysis. As the difficulty 

 offered in this respect was already seen when the mineral was 

 discovered at the locality, a comparatively large number of 

 specimens on which the mineral occurred were collected. From 

 these the diminutive crystal strips have been detached and 

 examined. Many of the individuals proved to be so intimately 

 associated with minute scales of graphite, feldspar splinters etc., 

 that they could not be used, but had to be rejected. The whole 

 quantity of pure material which could thus be obtained, amounted 

 after having been dried at 110°C., only to 0,0970 gr. On this 

 small quantity the analysis has been performed by Mauzelius 

 with the foUoAving result. 







Molecular ratios 





SiO, 



52,2 



0,864 





3,07 



Ah_ 0, 



2,7 



0,072 







FeO 



0,6 



0,008 > 



0^559 



2,00 



MgO 



19,1 



0,473 







Na^O 



11,5 



1,8 



0,122 

 0,029 ' 

 0,113 . 



0,264 



0,94 



Li^ 



3,8 







Loss 



8,7 



0,483*) 





1,72 





100 





Calculated as H., 



0. 









On the assumption that the loss in the analysis is water, 

 the chemical formula of the mineral лУоиЫ consequently be 



ZSiO.,-2 Mg . (Ä% , iVa 2 , Li., )0-2H.,0 

 or (Mg OH)^^ [K, Na, Li] Si. O^ + H.,0. 



The loss in the analysis can hardly be anything but water 

 or fluorine, most probably both. As, however, hydroxyl and 

 fluorine often play the same part in minerals, replacing each 

 other isomorphously, and as, moreover, these two substances 

 have nearly the same molecular weight, it is of but little im- 



