an uncommon Species ofTLeol'ite. 3^5 



by a gentle heat ; and the faline mafs left was pnt into a cru- 

 cible,\nd heated flowlv to rednefs. A white fall remaine<l 

 in the crucible, which' weighed 19 grains. To free this fait 

 from any remains of earthy falts which might be mixed with 

 it, I diffolvcd it in a fmaU'quantiiy of water, added fome car- 

 bonate of ammonia, and boiled the mixture for a few mi- 

 nutes, bv which means a flight earthy precipitate was thrown 

 down. 'This being feparated as before, the fait was agani 

 collefted by evaporation, and heated to rednefs. It no\V 

 weighed 17I grains, and was found on examination to be 

 pure fulphate of foda *'. 



By Mr. Kirwan's eftimation, 17I parts of dry fulphate of 

 foda' contain nearly eighc parts of alkali; confcquently, from 

 TOO parts of the zeolite there have been obtained by the pro- 

 cefs laft defcribed eight parts of foda. This refult corrc- 

 fponds with the former, in which the alkali was colieaed m 

 the ftatc of muriate of foda, as nearly as can be expefted iu 

 fuch experiments ; and it muft be remembered, that the pro- 

 portions of the component parts of neutral falts are not af- 

 certained with precifion. 



According to the different experiments now detailed, 100 

 parts of this zeolite contain, 



Silex (No. 1.) - - - 5^ '5 



Lime (No. 3 and 4.) - - S^' 



Argil (No. 3.) - - - '5 



Oxide of iron (No. 2.) - - *5 



Soda, about _ _ - - 8-5 



Carbonic acid and other volatile matter 5' 



. . . ^ 

 with fome traces of magnefia and muriatic acid. 



The ftone which has now been defcribed rcfembles fome 

 of the varieties of tremolite mentioned by SaufTure t, in the 

 property of giving a phofphoric light by friclion. Its Iprcific 

 gravity alfo is fomcvvhat greater than' that of the orchnary 

 kinds of zeolite, as ftated by mineralogifts. Except'mg m 

 theie pajliculars, however, it has the principal charattcrs of 



• The experiments which fhowtd that foda was the alkalne bafis of 

 t\\\i fait, h,v£ not been ftated here ; bcciufe I foVmcrly gave a de'cnpnon 

 of the methods ulcd in examining the fame ralt.in tlie paper on whinO-mc 

 and lava. pu!.[i(hed in part i. vol. v. liJmb. Trarilac. To avoid iinno- 

 cclfaiy itpctition, I hcg leave 10 refer to t!:at paper, both with regard 

 to tlie manner in which 1 endeavoured to determine t e purity of the laliiic 

 matter, and alfo that of the filex and fome olhcr CJlths. 



t Voy..£fi dan* le» Alpcs, 1913. ,. 



a zcoliic . 



