ANALYSIS OF LAftTJLITE. 21 



dodecaedron. The faces of the octaedron make an angle of 

 109 a 28' lfi": those of the dodecaedron an angle of 120°: 

 and the former cot these at an angle of 144° 44' 8". Be- 

 side these severtl smaller faces were observed, which were 

 not easy to determine, because the specimens were not 

 very distinct. 



It is not uncommon to meet with flattened quadrilateral Varieties of 

 prisms, the faces of which form angles of 101° 32' and form * % 

 78° 28'; angles that occur in several minerals, particularly 

 in the calcareous spar. At the extremities of these prisms 

 were faces in greater or less number, which we could not 

 ascertain. 



As to its contexture, we could not find it split decisively Not fissile. 

 10 any direction. 



With respect to its crystallization it can be compared only Resembles th« 

 with the spinelle, with which Mr. Haiiy classes the ceylanite Spme 

 or pleonast. As analysis informs us too, that it resembles 

 it in its constituent parts, we must consider them as similar. 



The following is a comparative analysis of them. 



Of the ~ ei . 

 Of the ceyla- °J£? 

 Of the spinelle by spinelle nite by . . 

 V.u,u.lta. by Kla- Coll*. ^^ 

 proth. Desco- , ff 

 tils. 



Alumine 860 82*47 745 68 66 Comparative 



Magnesia 8'5 S'7S 8.25 12 18 anal y s,s - 



Silex 15*5 2 10 



Lime ' 0*75 2 



Oxide of iron • • 1'5 l() 2'5 



Oxide of chrome 5 25 6*18 



We find that alumine united with magnesia must be con- 

 sidered as the essential part of the mineral. 



As Mr. Bernhardi took upon himself to describe the cha- 

 racters of the lazulite, Mr. Trommsdorff attended more 

 particularly to the analysis. He proceeded as follows. i ts analysis. 



A. A hundred grains of siderite strongly calcined in a calcined 

 covered crucible lost 5 grains of their weight. The fine 

 blue colour had disappeared, and was changed to a yellowish 

 white. 



B. The 



