Intelligence and Miscellaneotis Articles. 301 



hope will possess some interest, as they settle the chemical formula 

 of several substances on which the opinions of scientific men do not 

 agree. 



To establish the normal composition of the minerals formed in the 

 manner of sulphur, I suppose that wherever a certain proportion of 

 fluorine or chlorine is met with in these minerals, these bodies only 

 exist there still because the reaction which should have converted the 

 chlorides and fluorides into oxides has only been imperfectly eff'ected. 

 But if we complete, by calculation, the substitution of oxygen for 

 the chlorine and fluorine, converting the whole into oxides, we arrive 

 at a simple formula. 



For the topaz, these suppositions, already sufficiently justified by 

 the discussion of the analyses of this mineral due to M. Forch- 

 hammer, have just been confirmed by some unpublished researchs 

 of my brother and M. Fouque. The topaz would thus be as- 

 similated to a silicate of alumina, in which the oxygen of the base 

 would be in relation to that of the acid in the proportion of 4 to 3. 



When the mineral contains, together with fluorine, a body which, 

 like silicium, and especially boron, is capable of being sublimed with 

 it during calcination, we may, by taking the most probable hypo- 

 thesis, restore to the mineral its primitive composition. I find an 

 example of this in the tourmaline. It is well known that mineralo- 

 gists are far from being agreed as to the formula which should be 

 attributed to this substance. In his last supplement, M. Rammels- 

 berg proposes three diff"erent formulae, which have scarcely anything 

 in common, to represent the diff^erent varieties of a mineral so homo- 

 geneous in its crystalline form and in its properties. I have no space 

 here to discuss the numerous analyses of tourmaline contained in the 

 same volume ; but if we suppose that the loss by calcination, which 

 varies from TS to 3'8 per cent., is owing to a disengagement of 

 fluoride of boron, and calculate from this datum and from the boron 

 determined, the corresponding quantity of boracic acid, we obtain the 

 following numbers for the oxygen of the four constituent elements 

 of the mineral : — 



A remarkable concordance, which, I think, justifies us in attributing 

 a single formula to all the varieties of tourmaline, according to 

 which the oxygen would be distributed amongst the four elements 

 in the following proportions : — 



4: 6: 15 : 18. 



ITiis formula would certainly not apply to the last group of tour- 

 malines of M. Rammclsberg, the tourmalines containing lithia. 

 But the presence of this base may throw a light upon the causes of 

 this anomaly ; the litliia must in great part be carried ofl" by the 

 fluorine, and this supposition is confirmed by the researches of 

 MM. H. Saintc-Claire Dcville and Fouque. Moreover, when the 

 same hypothesis is applied to analyses of tourmalines which doubt- 



