traces on paper. When well washed in boiling water, it only communicated to that liquid imperceptible traces of alkali.

When exposed to a strong heat, it could not be incinerated. When treated in a crucible with potash, it gave some indications of the presence of a cyanide. This charcoal was almost completely burned by means of nitre. The alkaline mass was dissolved in water, and an excess of muriatic acid poured into it. Ammonia was then added to the filtered liquid, which produced a precipitate weighing 0.4 gramme. It was phosphate of lime; for when dissolved in a little nitric acid, the subacetate of lead occasioned a precipitate, which, when well washed, was fused before the blowpipe into a crystallized bead of phosphate of lead. Subcarbonate of soda being added to the liquor from which the phosphate of lime had been precipitated by ammonia, precipitated, when assisted by heat, about three centigrammes of carbonate of lime.

The ashes of rice, then, consist almost entirely of phosphate of lime. On appreciating as exactly as possible the results of the comparative analyses of Carolina and Piedmont rice, I consider the approximate constituents of each to be as follows:

Memoir on the Sodalite of Vesuvius. By M. Le Comte Stanislas Dunin Borkowski.

(Presented to the Academy of Sciences, Oct. 28, 1816.)

M. Ekeberg was the first who analysed a mineral from Greenland which contains 25 per cent. of soda. Dr. Thomson repeated this analysis, gave a mineralogical description of the mineral, and

made it known in his excellent memoir as a new species, under the name of sodalite.* No other locality of sodalite has hitherto been observed; but I have been lucky enough to find it on the slope of Vesuvius, called Fosso Grande, which may be fairly considered as the great repertory of the volcanic riches of Vesuvius. Sodalite appears due to the ancient eruptions, which have furnished mineralogy with nepheline, meyonite, and idocrase; but it is very far from being so common as these species. This is no doubt the reason why it had not been observed by the skilful observers who have examined that celebrated country. Notwithstanding considerable search, I could meet with only a single specimen on the spot. Another was afterwards given me by the guide Salvatore. The following observations were made upon these two specimens :

External Characters.

The sodalite of Vesuvius is greyish-white. It occurs in round grains, and crystallized in the form of six-sided prisms, terminated in a point by three faces placed alternately on the lateral edges. The crystals vary in size, and I possess one which is an inch long. The surface of the crystals is smooth, and rather irridescent.

1

External lustre shining, resinous. Internal lustre vitreous. Cross fracture perfectly conchoidal; principal fracture foliated; but it is difficult to determine the cleavage. Translucent. Fragments indeterminate; sharp edged.

It is semihard, yielding readily to the file; easily frangible. Specific gravity, 2·89.

Chemical Characters.

The fragments of the sodalite of Vesuvius, when put into nitric acid, do not lose their lustre while in the acid; but after they are taken out, they soon become covered with a whitish crust. When put in powder into muriatic acid, they form a jelly. Before the blow-pipe, it melts, without addition, but with difficulty.

Position.

The sodalite is found in calcareo-talcose quangue, accompanied by pyroxene, green pumice, and a substance crystallized in small six-sided tables, called by Werner icespar.

The mineralogical characters which I have just given presented an unknown substance; but they were far from ascertaining its real nature. Nor could crystallography serve to determine the species; for the form of the crystals, being a six-sided prism, terminated by three-sided pyramids, with angles of 120°, gave for a primitive form the rhomboidal dodecahedron; but this primitive form, being

* This statement would require some correction. When I made my analyses, I was not aware that the mineral had been examined by Ekeberg. Ekeberg, I believe, never published any thing on the subject. He merely sent the numerical account of his analyses to Mr. Allan, in whose possession I saw it.-T.

common to several different species, ceases on that account to be distinctive. It was necessary, therefore, to have recourse to chemistry; and the result of my analysis completely answered my intention. Chemical Analysis.

A.-25 decigrammes of sodalite in fragments were kept at a cherry-red heat in a platinum crucible for half an hour, without losing any weight. They became milky in their aspect. The angles which touched the sides of the crucible had undergone a commencement of fusion.

B.-1. 4 grammes of sodalite reduced to a fine powder were mixed with 10 grammes of muriatic acid diluted with five parts of distilled water. The stone was attacked in the cold. On a gentle ebullition, the solution assumed the form of a stiff yellow jelly, which I collected with much care upon a porcelain capsule, and evaporated the whole to dryness. Towards the end of the evaporation, care was taken to stir the jelly, that the drying might be gentle and equal. When the whole was reduced to powder, it was diluted with water, and the residue washed till the liquid ceased to affect nitrate of silver. This residue, being heated to redness, weighed 17.25 decigrammes. The filter had increased in weight 0-25 decigrammes, which makes the total weight of the residue 17.5 decigrammes.

To convince myself that this residue was silica, I heated it for half an hour, with five grammes of caustic potash, in a silver crucible. The mixture fused; and, being taken from the fire, I poured distilled water on it while still hot. When well diffused through the water, I poured muriatic acid on it, which dissolved it completely. This solution was evaporated to dryness. The silica obtained, after being washed and heated to redness, weighed 17 decigrammes. The loss of 0.5 decigramme was owing to not having weighed the filter; for the liquid employed to wash it was neither precipitated by caustic ammonia nor by carbonate of ammonia.

2. The acid liquid from which the silica had been separated was precipitated by pure ammonia. A very white, bulky matter was obtained, which was separated by the filter. After being washed, it was boiled, while still moist, in caustic potash. The whole was dissolved, except a little brown matter, which was separated by the filter. The alkaline liquor being neutralized by muriate of ammonia, a copious precipitate fell, which, after being washed and heated to redness, weighed 6.75 decigrammes. It possessed all the properties of alumina.

3. I poured carbonate of ammonia into the liquid from which the alumina had been separated. Next day I found a precipitate, which, having been washed and heated to redness, weighed 2.75 decigrammes. The residue dissolved in sulphuric acid was evaporated to dryness, and treated with cold water, which dissolved the whole. This solution was concentrated by evaporation, and set aside

for spontaneous crystallization. As it refused to crystallize, and had not the taste of sulphate of magnesia, sulphate of potash was added to it, on which crystals of alum were formed. The precipitate obtained by the carbonate of ammonia was, therefore, alumina, which must be added to the sum obtained by the preceding experi

ment.

4. As there was a matter attached to the rod from which the 2.75 decigrammes of alumina were obtained, I poured muriatic acid on it, in order to obtain this matter. Brilliant scales were detached, which, when collected on the filter, and dried, had so strong a resemblance to boracic acid, that I thought at first that I had obtained that acid; but I soon satisfied myself that it was silica. It weighed 0.25 decigrammes.

5. The brown deposite of the second experiment, which weighed 0.25, was treated with sulphuric acid, which dissolved the iron without touching the silica. The iron, being precipitated from the solution by ammonia, weighed 0.05 decigramme. This metal exists in such a minute proportion, that I think it belongs rather to the green pumice than to the sodalite. The O2 decigramme not attacked by the sulphuric acid possessed the characters of silica.

6. The silica, alumina, and iron obtained, not amounting to the weight of the stone analyzed, it was necessary to seek for the other constituents in the liquid from which the earths had been separated by the carbonate of ammonia. The liquid, in consequence, was concentrated; sulphuric acid was added, to drive off the muriatic acid, and convert the same into sulphate. It was then evaporated to dryness, and exposed to a red heat, to drive off the sulphate of ammonia and the excess of sulphuric acid. The matter obtained weighed 22.5 decigrammes. It was dissolved in water; the solution was concentrated, and set aside. Some needle-form crystals of sulphate of lime were deposited; but the quantity was so small that they cannot be estimated. The liquid had crystallized confusedly in small crystals; and as it precipitated the solution of platinum, I thought at first that it was sulphate of potash. But when the crystals were redissolved, the liquid furnished, by spontaneous evaporation, six-sided prisms, which effloresced in the air, had a cooling taste, and did not precipitate platinum. They had, therefore, all the characters of sulphate of soda; and as the sulphate of soda obtained by calcination weighed 22.5 decigrammes, it contained 11 decigrammes of pure soda. The precipitate obtained by the platinum solution is owing to the presence of a small quantity of potash which is mixed with the soda.

The mineral analyzed, then, consists, supposing it to be 40 parts, of the following ingredients :

The great quantity of soda which I obtained made me immediately suspect that the substance analysed was a sodalite. This suspicion was confirmed when I compared my analysis with those of Ekeberg and Thomson. The following table exhibits their results :

These analyses differ from mine merely in my having found a little potash mixed with the soda. The loss of 3.76 which I had in my analysis corresponds with the three parts of muriatic acid obtained by Dr. Thomson, which it was impossible for me to perceive, as I had employed that acid in my analysis. The external charac ters of the sodalite of Greenland do not differ essentially from those which I observed in the sodalite of Vesuvius; for the six-sided