proceed from the immediate direction of some human agent; and since there is no communication with such an agent concealed within the chest, or in a room adjoining, it must proceed from the immediate direction of the exhibitor himself.

Nevertheless, the operation of this directing force, or in what secret manner the exhibitor directs the arm and hand of the automaton when raised, yet remains to be, explained. M. de Kempelen once threw out a hint, that the chief merit of his invention lay in the successful manner in which he deceived the spectators; by which hint he seemed to imply not only that the exhibitor does interfere in an unperceived manner in directing the arm and hand of the automaton when raised, according to the varying circumstances of a game of chess; but that the mode of such interference is very simple. In fact, when the arm and hand are raised and prepared to act by the operation of the moving force already explained, the action of a wire, or piece of catgut, not much thicker than a hair, would be sufficient to guide them in any direction; which action, from the delicacy of the medium used, might be communicated in a manner wholly unperceived by the spectators *.

Probably the precise time and instrument of communicating, this action, which are circumstances systematically kept secret, will never be discovered; and the conception of them reflects the highest honour upon the ingenuity of the inventor. To construct an arm and hand

*There can be little doubt that the peculiar action of the automaton, by which the knight is made to touch each of the sixtythree squares of the chess-board in turn, depends upon the action of machinery alone, without any interference of the exhibitor, except in previously winding up the works. The motions of the head of the figure, and its tapping on the chest, are a kind of hors d'œuvre.

capable

capable of performing the ordinary functions of those parts, would be of itself sufficient to secure the reputation of an artist; but to make the same arm and hand almost counterparts of living members in a reasoning agent, displays a power of invention as bold and original, as any that has ever been exhibited to the world.

Letter of M. GILLET LAUMONT, to the Editors of the Annales de Chimie, on the Subject of Lithion. Dated April 6, 1818.

M. SUDENSTIERNA informs me, that M. Asfredson had not entirely compleated his analysis of lithion when he wrote to me that he had extracted three per cent. of it from the petalite, and that he now finds it contains five per cent.; which approaches nearer to the analysis of M. Vauquelin, who obtained seven per cent. from very pure specimens.

M. Sudenstierna further informs me that M. Asfredson has procured eight per cent. of litheon from the triphane, doubtless from Abo. This further discovery will enable chemists to procure the new alkali with more ease, the triphane being less rare than the petalite; and M. Leonbard has just discovered some of it in the Tyrol, which has yielded to M. Hisinger's analysis six per cent. of an alkali that he considers as potash, but which may probably prove to be lithion.

Memoir on the Sorbic Acid, and its different Combinations. By M. HENRI BRACONNOT.

From the ANNALES DE CHIMIE ET DE PHYSIQUE.

IN my experiments to obtain the malic acid in a

state of perfect purity, I hoped to succeed by means of the malate of zinc, which, as Scheele has observed, gives very fine crystals, and of which I prepared a considerable quantity; but I soon perceived that the acid which I supposed to be the malic in combination with the zinc, possessed properties very different from those of this acid. In the mean time I learned that Mr. Donovan had described, in the Philosophical Transactions for 1815, a new acid which he names the sorbic, on account of its being obtained from fruits of the sorbus species, and which I doubted not was the acid of the salt of zinc that I had obtained. Though Mr. Donovan had published his observations before mine, I resolved to continue my experiments and to repeat his.

To obtain the sorbic acid, Mr. Donovan precipitates the juice of the mountain ash berries (sorbus aucuparia). by acetate of lead, washes the precipitate by cold water, then treats it with successive portions of boiling water, to collect, by cooling, the crystals of sorbate of lead, from which the acid is afterwards separated by sulphuric acid and sulphuretted hydrogen.

This process appearing to me defective on account of the small quantity of product obtained, I proceed in the following manner: take mountain ash berries a little before they are ripe, bruise them in a marble mortar and press out the juice; boil it in a bason, and add carbonate of lime till effervescence ceases, evaporate the liquid to the thickness of a syrup, removing the scum as it

rises; a copious precipitate of sorbate of lime falls down as a granulated salt, which must be stirred to prevent it from adhering to the bason, and after standing some hours to cool, pour the syrupy liquid from off the sediment of sorbate of lime, rinse the latter with cold water, and press it in old linen to dry it. This salt is slightly yellow, shewing it to be still impure; boil it for a quarter of an hour with its own weight of crystallised carbonate of soda dissolved in water, which will produce a sorbate of soda soiled with a little red colouring matter: to remove this latter, boil it a few minutes with lime-water or milk of lime, and the liquor will now be a clear colourless solution of sorbate of soda. Pass a current of carbonic acid through it to throw down the lime in solution, then add sub-acetite of lead, which will produce a very white precipitate of sorbate of lead, from which the sorbic acid may be separated by heating with dilute sulphuric acid.

This acid may also be obtained by partially decomposing the impure sorbate of lime by sulphuric acid, which will produce a super-sorbate of lime easily crystallizable. Dissolve this salt in warm water, add weak sulphuric acid till no more precipitate separates, evaporate the liquor, and re-dissolve in alcohol to get the sorbic acid in solution, and exclude the sulphate of lime; or else the sorbic acid may be at once separated by sulphuric acid from the neutral sorbate of lime; only in this case the colouring matter must be first got rid of, by dissolving the sorbate of lime in boiling water and again'crystallising, which gives a very white salt.

The sorbic appears to be a common vegetable acid: I have found it in sensible quantity in verjuice, which also contains much citric acid, as Scheele first observed. To separate these acids from each other as they exist in the

juice of unripe grapes, saturate the boiling juice with chalk; the citrate of lime being the least soluble, falls down first; decant the supernatant liquor, and evaporate it, when it will yield a granulated deposit of sorbate of lime, which may be then decomposed by sulphuric acid.

The sorbic acid is obtained pure by either of these processes. It is uncrystallisable and deliquescent, and it gives out the smell of scorched sugar in burning.

General Character of the Sorbates.

The sorbates have all an analogy with the tartrites in their property of combining with an excess of acid which diminishes their solubility. But as the sorbic acid is itself uncrystallisable, it must of necessity form more soluble salts than the tartaric acid with the same bases; so that it may be taken as a general rule, that where the latter acid forms with any base a salt nearly insoluble, the sorbic acid forms with the same base a salt that will crystallise; and where the tartrite will not crystallise, the sorbate of the same base is still less disposed to crystallise.

In the sorbates, 100 parts of sorbic acid saturate so much of each base as contains 11 of oxygen. According to Berzelius 100 parts of tartaric acid saturates a base of 11.94 oxygen, the same quantity of citric acid saturates 13.588, of acetic acid 15 43, and of oxalic acid 22.062.

The super-sorbates contain exactly twice as much acid as their neutral salts. All these salts swell more or less in burning, and have a disposition to adhere to the vessels in which they crystallise.

Sorbate of Potash, Soda, and Ammonia.

These salts have been examined by Mr. Donovan, they are very soluble, and will not crystallise unless with an excess of acid.

Sorbate