488 
THE GKOLOGIST. 
liscd in very bard hexagonal laminae ; but it was not emerald, for by 
analysis it was proved deficient in silica. 
It will be perceived from vpbat has just been related, that fluoride 
of aluminium decomposes silica to form fluoride of silicium and stauro- 
tide. And, also, that in the same manner, fluoride of silicium in con- 
tact with aluiniua gives fluoride of aluminium and staurotide. This is 
the reusou why our smelting crucibles and other metallurgical apparatus 
are often entirely transformed into a rough mass of crystals, almost all 
of which consist of the mineral species staurotide. The preceding 
observations explain also how volatile fluor-compounds in the presence 
of argillaceous substances may become the means of obtaining crj^stals 
of substances which are perfectly insoluble or infusible at the tem- 
peratures at which these fluor-compounds operate. This remarkable 
iufl^uenee of metallic fluorides in the production of crystallized mineral 
species must have played a certain part in nature — more particularly 
in the production of crystallized minerals in lodes and veins. At least, 
such is the opinion of M. H. Deville ; and M. D.iubrce seems already 
to have arrived at this conclusion in his beautiful memoirs on the 
metallic fluorides. At the end of their paper, Messrs. Deville and 
Caron state that they have likewise artificially obtained rutile, or 
titanic acid, an interesting mineral species, in decomposing a fusible 
titanite by silica. Titauite of protoxide of tin was preferred for this 
experiment. During the operation, the authors remarked dissolved in 
the metallic tin, a brilliant substance, crystallised in wide metallic 
laminaj, easily separated from the tin by dissolving the latter in hyro- 
chloric acid, and which turned out to be an alloy of iron and tin, con- 
taining equal proportions of each. 
It was staled several months ago, that the immense work of a tunnel 
under the Alps between Modaue and Eardoneche had commenced. We 
have now to record some interesting facts, which might perhaps never 
have been discovered but for the peculiar methods employed in this 
colossal operation. The crest of the mountain attains so great a height 
that the sinking of shafts, which is generally done in order to begin 
boring at several points at once, was in this case out of the question. 
Hence the tunnel could only be worked at its extremities, so that the 
labour by the ordiuary process of working could not be accomplished in 
less tlian thirty-six years ! Then, how was a depth of gallery of one or 
two miles, or more, and having but one orifice, to be aired ? These 
were very serious obstacles. 
MM. Elie de Beaumont and Angelo Sismonda having examined the 
raouutain geologically between Modane and Bardoneche, found it com- 
posed of micaceous sandstones and schists, quartzite, gypsum, and lime- 
stone, all easy to blast, the quartzite alone excepted ; but the stratum 
of this is not likely to be very thick. The other difficulties alone, 
therefore, remained, and these were at length overcome by three 
Sardinian engineers, MM. Sommeiller, Graltone, and Grandis, who 
proposed to turn the abundance of water, for which the locality is 
remarkable, to account by applying it to a peculiar system of perfora- 
tion and ventilation, which we will now endeavour to explain. The 
first apparatus imagined by these gentlemen consists of an hydraulic 
