Mr. Connel's Analysis 0/" Levyne. 41 



dinger*. At the same time he transmitted a small quantity of 

 the mineral to Berzelius for chemical examination ; and soon 

 afterwards, in a paper inserted in the Transactions of the Royal 

 Swedish Academy for 1824f, Berzelius gave the following as 

 the constituents of the mineral : 



Silica 4.8-00 



Alumina 20*00 



Lime 8*35 



Magnesia 0*4 



Potash 0*41 



Soda 2-75 



Water 19*30 



99-21 

 Berzelius accompanied his analysis with a statement that 

 he considered the mineral to be merely a chabasite. From a 

 subsequent explanation of Sir David Brewster J, however, there 

 is every reason to believe that Berzelius had not distinguished 

 the levyne from some chabasite which accompanied it in the 

 same specimen, and that in reality a mixture of both minerals 

 had been subjected to analysis, a mistake which is not very 

 surprising, considering that levyne was at that time entirely a 

 new mineral, and its aspect consequently little known. 



Had not a reasonable doubt been thus thrown on the iden- 

 tity of the subject of Berzelius' s analysis, any further exami- 

 nation of the chemical nature of levyne would have been a very 

 superfluous task; but under the circumstances, it became 

 very desirable that a new analysis should be executed of un- 

 doubted crystals of this mineral. 



The specimens examined were partly furnished me by the 

 kindness of my friend Mr. Robert Allan, and partly in my 

 own possession. They were all of Irish locality §. The cry- 

 stalline form was quite distinct, and exactly that described by 

 Mr. Haidinger. I regret that it was not in my power to ope- 

 rate on larger quantities of materials than those employed. 



To determine the specific gravity of the mineral, the largest 

 portions of the crystals separated were selected, the quantity 

 so selected amounting to only 9*3 grains. Their specific 

 gravity was found to be 2*198 at 55° Fahr. The mineral 

 by ignition gave off water, the proportion of which in a mean 

 of two trials, giving almost exactly the same result, amounted 

 to 19*51 per cent. 



To determine the proportion of the other constituents, 

 10*28 grains of the crystals in impalpable powder were treated 



* Edinburgh Journal of Science, vol. ii. p. 332. f p. 356. 



± Edinburgh Journal of Science, vol. iv. p. 316. 



§ In addition to the other known localities of this mineral, I may 

 mention that I have found specimens of it in the island of Skye. 

 Third Series. Vol. 5, No. 25. July 1 834 . G 



