gave out, when placed on a burning coal, an empyreumatic smell of animal matter. It is this animal substance, doubtless, which imparts to the aqueous solution its opaline appearance.

After the liquor had been filtered, in order to separate the red matter, it was saturated with carbonate of lime, the precipitated sulphate removed by a second filtration, and the fluid evaporated to the consistence of a thick syrup. This did not crystallize; it seemed to be more saccharine than the gum was prior to the operation, but was insoluble in alcohol. This gum, therefore, is not of the same kind as that obtained from starch, by the corresponding treatment with sulphuric acid.

Treated with nitric acid, it yielded much oxalic acid, and a small quantity of yellow bitter matter, but no mucic acid: this proves that it is not a true gum. A gramme of it burnt in a platinum crucible left a centigramme of ash, containing phosphate of lime, iron, and a particle of silica.

M. Vauquelin concludes, that this curious substance did not exist ready formed in the vesou, but that it was produced from the sugar contained in it.

་

ARTICLE XIII.

On a peculiar Sulphate of Alumina. By Richard Phillips, FRS. L. & E. &c.

INTENDING Some time since to obtain a solution of sulphate of alumina in a state as nearly as possible approaching to saturation, I decomposed some alum by means of carbonate of soda, and after washing the precipitated alumina, I put it, while moist, into sulphuric acid, moderately diluted with water. Although the acid appeared to have taken up as nearly as much alumina as it was capable of dissolving, I nevertheless added alumina occasionally, until at last it remained floating in the solution.

I now separated the alumina undissolved, and filtered the solution, the specific gravity of which was very considerable; upon mixing a small quantity of it with water, I was surprised to find that it became turbid, and nearly as much so as when muriate of antimony is decomposed by it; indeed the alumina of a single drop of the solution was apparent in a pint of water. As far as I am acquainted with the properties of alumina, a sulphate decomposable by water has not been before observed; and it may be remarked that it is an additional point of resemblance between an earth and metallic oxides.

After the solution had been filtered, I observed that a deposit was almost immediately formed in the bottom of the bottle in which it was kept; this was separated, and a further

and placed in water of the temperature of 1600 to

quantity was obtained; indeed I found that during several months, the solution continued depositing, but the substance had not in any degree a crystalline form. Another property of this solution is worthy of notice: if some of it be put into a tube, 170°, and probably even lower, it becomes opaque and thick in a few seconds; if, however, the tube and its contents be kept at the ordinary temperature of the air for several days, the precipitate is gradually redissolved, and the solution regains its transparency. It appears extremely singular that this solution should have required so long a time for its production, and perhaps still more so that the peculiar sulphate of alumina in question was not deposited as quickly as it was formed; yet I did not observe any disposition to deposit until after the removal of the excess of alumina floating in the solution. Not anticipating the spontaneous deposition which I have described, I did not take the specific gravity of the solution at its greatest density; but after it had continued depositing for several weeks, I found the specific gravity of the solution exceeded 1·120.

Although the solution of sulphate of alumina continued affording a deposit for many months, yet it did not appear to suffer any change of composition, for water added to it at this period continued to occasion precipitation, which it probably would not have done, if the deposit consisted of alumina combined with less acid than when in solution, for the excess of acid which must have remained in solution, would probably have prevented the precipitating action of the water.

As metallic oxides which are precipitated from acid solutions by water, usually contain a portion of the acid which held them in solution, there could be no doubt that the precipitate formed in this sulphate of alumina by water was a subsulphate, and I found it to be so, but I have not yet had leisure to determine its composition.

It is well known that it is extremely difficult to deprive alum of the whole of its sulphuric acid, and I found that alumina, even when precipitated from solution by excess of ammonia, and ignited, gave a precipitate with muriate of barytes when redissolved in an acid. It appeared to me, therefore, a question to be decided what quantity remains in combination with the alumina. I dissolved 1000 grains of alum in water, precipitated the alumina by carbonate of soda, and washed it with distilled water until it ceased to afford sulphuric acid, as determined by nitrate of barytes. I then dissolved the alumina in nitric acid, and added nitrate of barytes as long as precipitation occurred; the sulphate of barytes when dried weighed 24 grains, consequently the precipitated alumina contained 8.1 of sulphuric acid; and as 1000 of alum yield about 110 of alumina, I shall, in the experiments which I am going to state, deduct 7.36 per cent. from the precipitates of alumina, considering it as sulphuric acid.

To 736 grains of this solution, water was added as long as precipitation took place; the precipitate was dried by exposure to the air, and weighed 40 grains; 100 grains of the solution would, therefore, give 5.43 grains. I repeated this experiment with 1020 grains of the solution, which yielded 52.5 grains of precipitate dried as before; 100 would consequently have afforded 5-14 grains, giving a mean of 5.23 grains of subsulphate of alumina from 100 grains of the solution.

To determine the quantities of sulphuric acid and alumina which the precipitating sulphate of alumina contained, muriatic acid was added to 392 grains; this acid was of course employed to prevent the action of the water; nitrate of barytes was added, and 61 grains of ignited sulphate of barytes were obtained, equivalent to 15.56 per cent.; this experiment was repeated with 205 grains of the solution, and 31.9 of ignited sulphate of barytes were procured, giving also 15.56 per cent. As 118 of sulphate of barytes are equivalent to 40 of sulphuric acid, 15.56 will indicate 5.27, and consequently 100 grains of this solution contain 5.27 of sulphuric acid."

To 633 grains of the same solution, with which a little muriatic acid had been mixed, solution of carbonate of soda was added, until it was slightly in excess. The precipitated alumina, after being washed and ignited, weighed 36 grains; 100 grains of the solution would, therefore, have yielded 5.68 grains: this experiment was repeated with 625 grains of the solution, and 37 of ignited alumina were obtained; 100 of the solution would, therefore, have afforded 5.92, giving a mean of 5-8 of alumina for 100 of the solution. From this, however, for reasons already stated, we must deduct 7.36 per cent. which reduces it to 5.38. It appears then that 100 grains of this solution contain

Sulphuric acid.
Alumina.

5.27
5.38

According to Dr. Thomson, hydrogen being 1, an atom of sulphuric acid is 40, and of alumina 18; and as 5-27: 5·38 :: 40 :40 83 the sulphate of alumina of this solution would not appear to be reducible to a probable definite compound; but I have already mentioned that a deposit was formed in it which appeared to be the same sulphate as that held in solution, for water continued to decompose the latter.

The deposited sulphate, when dried by exposure to air, is in some places opaque, and in others transparent; and when in the latter condition, it has the appearance of horn. To ascertain its composition, I dissolved 50 grains of it in dilute muriatic acid, and added a solution of muriate of barytes; 38.5 of ignited sulphate of barytes were obtained; therefore, 100 grains would have given 77 grains, equivalent to 26-10 of sulphuric acid. To ascertain the proportion of alumina, 100 grains dissolved in dilute

muriatic acid were decomposed by carbonate of soda, and 28.8 of alumina remained after ignition. Deducting 7.36 per cent. from the alumina for sulphuric acid, this deposited sulphate of alumina very closely resembles that of the solution, and appears, therefore, to have been deposited without any decomposition. I have just shown that the solution consists of

From various considerations, more especially the constitution of alum, I am induced to differ from Dr. Thomson as to the weight of an atom of alumina. I shall take an early opportunity of returning to the subject; and, I think, I shall be able to show, contrary to his views, that alum contains a supersalt. Among other reasons for this opinion, I may state one experiment which I have very frequently repeated. If zinc filings be added to a solution of alum, they are gradually dissolved, but with sufficient rapidity to give out enough hydrogen gas to cause an explosion when a flame is presented.

According to my present opinion, an atom of alumina weighs 27, or one-half more than determined by Dr. Thomson. On this view the deposited sulphate of alumina which I have described will consist of

I have already observed that when this solution is mixed with water, it is decomposed; and I have some reason for believing that the sulphate of alumina which remains in solution is that which with bisulphate of potash forms alum, the precipitate being, as I have ascertained, and indeed already mentioned, a subsulphate of alumina.

ARTICLE XIV.

Extract from a Memoir on the Composition of the Alkaline Sulphurets. By M. Berzelius.

M. BERZELIUS Commences this paper with a history of the present state of our knowledge with respect to these compounds; and he then proceeds to detail the experiments which he has performed to elucidate the subject, beginning with

Experiments to determine whether the Sulphuret formed in the dry Way is a Sulphuret of the Oxide, or of the Metal.

If sulphuret of potassium can exist, it is evident it ought to be formed when sulphate of potash is decomposed; and after the solution of the compound in water, the nature of the result must depend upon the formation of a sulphuret of potash or potassium. To verify this, I made use of a small apparatus with an enameller's lamp, and so constructed that a current of hydrogen gas might be passed through it, while part of the apparatus was heated to redness by an argand spirit lamp. In this part one gramme (15.444 grains) of neutral sulphate of potash was introduced. This salt did not suffer any change for some time, but when the heat was raised, small red points were seen in parts which readily increased, and water was formed. The matter became black, and fused. The operation was continued as long as the gas introduced appeared to produce water, which was collected in muriate of lime. The salt, when cold, was of a fine cinnabar red colour; it had lost 0-315 gramme, and the water produced weighed 0.335 gramme. The red mass was easily dissolved by water, which became of a very light yellow colour. It deposited some silica yielded by the glass, muriatic acid evolved sulphuretted hydrogen with effervescence, and the solution was rendered slightly opaque by a little sulphur. Decomposed by muriatic acid, it gave with muriate of barytes 0.157 gramme of barytes, corresponding to 0.108 of sulphate of potash; the 0-335 gramme of water produced contain 0-298 of oxygen; but the sulphuric acid in one gramme of sulphate of potash contains only 0-275, and the potash 0.092 of oxygen. Then if it be remembered that there remained at the close of the experiment one-tenth of the salt which did not appear to have been decomposed, it will appear that about two-thirds of the potash were reduced to potassium, and that the remaining one-third combined with the glass when it lost its sulphur, one portion of which combined with the potassium, and the other was carried off by the hydrogen in the state of a white