longer. From the very first, a peculiar penetrating substance was disengaged, having a peculiar smell (which is not similar to that of prussic acid). When a burning body is brought into contact with this substance, it takes fire, and burns with a light white flame mixed with blue, which may be extinguished by again putting the cover over the crucible. This gas is either the cyanogen discovered by Gay-Lussac, or prussic acid; for if a paper moistened with liquid ammonia be held over it for a while, it gives with an acid solution of oxide of iron excellent prussian blue; but I have not made any more accurate experiments on it. The mass in the crucible concretes together, melts, assumes an appearance very similar to graphite, and shows here and there, particularly along the fracture, a quantity of small metallic specks, similar to iron with a silvery lustre. When at this period, the fire is raised, by blowing to a white heat there comes over at last another gas, the bubbles of which as soon as they pass through the melted graphite-looking matter take fire in the air of their own accord. A white flame and weak explosion characterize this combustion. If a polished plate of steel be held. over the crucible at this period, it becomes covered with subtile flocks of a whitish grey colour, which act as an alkali upon turmeric paper. As the gas is always distinguished by the peculiar smell already mentioned, and as it always forms prussian blue when treated with liquid ammonia and an acid solution of oxide of iron, it must consist of cyanogen mixed with potash, or of prussic acid gas containing potash.*

Sect. 3.-The black, graphite looking matter in the crucible is now allowed to cool, it is taken out, rubbed down to powder, and set to digest in alcohol. After some time, the liquid is passed through the filter, and a new portion of alcohol poured upon the powder. This digestion of new alcohol is continued till the liquid ceases to alter the colour of solutions of iron. The alcoholic solution is usually colourless and transparent, though sometimes it has a blood red colour, owing to the presence of a portion of anthrazothionate of iron, from which it is easy to free it by the addition of a few drops of the alcoholic solution of

When the experiment is repeated in proper metallic tubes (which better keep off the action of atmospherical oxygen), there can be no doubt that potassium will be obtained at a lower temperature, and with greater ease, than by the method hitherto practised. Ought not this gas to be formed when potassium is heated in eyanogen? And ought not the hydrogen gas, which remains behind when the excess of cyanogen is absorbed by potash, to be ascribed to the decomposition of the water which the potassium in the cyanogen finds in the potash? Gay-Lussac found that when 48 or 50 volumes of cyanogen were combined with potassium, the remaining cyanogen, after it had been absorbed by potash (which always contains moisture), left a residuum amounting to 12 parts. But this is an equivalent to 24 volumes of hydrocyanic acid: hence it follows that the cyanogen under examination must have contained the third of its volume of hydrocyanic acid. When we consider Gay-Lussac's accuracy, and the excellent apparatus and reagents which he had at his disposal, so great an impurity in the cyanogen prepared by him, and contrary to his wishes, is very unlikely.

potash. The oxide of iron falls down, the red colour disappears, and the liquid when filtered becomes quite colourless. It contains pure anthraxothionate of potash, which by gentle evaporation is obtained in long channelled prisms terminated by foursided pyramids. It frequently crystallizes likewise in long, white, brilliant needles; it is much more soluble in boiling than in cold alcohol. Hence it frequently happens when a glass is half filled with a boiling hot saturated solution of this alcohol, that the salt during the cooling shoots up from the bottom of the glass in brilliant needles, the ends of which, in consequence of the adhesion of the liquid and its universal contraction, are some lines higher than the surface of the solution.

Sect. 4.-Anthrazothionate of potash has at first a hot taste, similar to that of radishes, but leaves in the mouth a cooling salt impression. In summer, when placed in dry air, it retains its crystalline form unaltered; but as soon as the air becomes in the least moist, for example, a little before the dew begins to fall, at six in the evening, it becomes liquid, and remains in that state till ten o'clock the following morning. When it has not become solid by that time, or at the latest by 11 o'clock, it is a proof that the atmosphere is moist; and we may predict with some probability that it will rain either on that day or the following one. For such an experiment the salt must not be exposed to the sunshine, but placed in the open air in a watch glass, and in a place moderately shaded. The remarkable delicacy with which this salt indicates the hygrometrical changes of the atmosphere, the readiness with which it gives out the water which it has absorbed when put into a dry place, fit it, I think, particularly for hygroscopical investigations; and if it were placed upon al scale properly balanced, it would form an excellent hygrometer. When exposed to heat in a glass tube, it melts quietly into a colourless and glass-looking mass, which on cooling congeals again in crystals. It is capable of bearing a much higher temperature without decomposition than prussiate of potash. However, when it is exposed to a red heat in certain metallic vessels, for example in silver, sulphuret of silver is formed which blackens the metal, and at the same time ammonia is produced. Sect 5.-Diluted sulphuric acid drives off the anthrazothionic acid undecomposed. We may, therefore, by its means easily obtain this acid in a free state in a proper receiver. The tedious and complicated method, therefore, which Porrett has given for obtaining it is quite unnecessary. Concentrated sulphuric acid drives off only a part of the acid undecomposed from the anthrazothionate of potash. The remainder undergoes a decomposition: an effervescence takes place; the sulphur separates in flocks, while carbonic acid gas and sulphurous acid gas fly off. But the most remarkable circumstance (which Porrett seems to have entirely overlooked) is that during this decomposition, which goes on rapidly when assisted by heat, no trace of azotic gar

appears, but the whole of it is converted into ammonia, which unites with the sulphuric acid. The ammonia may be detected by pouring an excess of concentrated potash ley upon the liquid after the chemical action is at an end. A very strong smell of ammonia immediately becomes sensible, especially if heat be applied. This observation, as will be seen below, was of great importance to me in forming a true notion of the nature of anthrazothionic acid.

Sect. 6.-Common concentrated muriatic acid likewise separates the acid from the salt without decomposing it; but it is not quite pure, being mixed with muriatic acid. Concentrated liquid chlorine prepared by pouring concentrated muriatic acid upon chlorate of potash decomposes the salt quite in the same way as concentrated sulphuric acid: the sulphur is precipitated in yellow flocks; but when the salt is mixed in the first place with chlorate of potash, and then muriatic acid poured over it, no sulphur separates, but it is completely converted into sulphuric acid, while the anthrazothionic acid is completely decomposed. At the same time muriate of ammonia is formed and carbonic acid is disengaged.

G Smoking nitric acid when poured upon the salt occasions a violent effervescence, nitrous gas and carbonic acid are given off, while the sulphur separates in substance.

Some vegetable acids, as, for example, tartaric acid, are capable of separating the anthrazothionic acid from the salt, especially when assisted by heat.

Sect. 7. In no one of the decompositions performed by means of the different mineral acids, though they were often repeated by me, have I been able to perceive so much as a trace either of prussic acid or of cyanogen in a free state, though I examined both the liquids that came over into the receiver and those which remained behind in the retort. No prussian blue was ever formed when they were treated with ammonia and an acid solution of oxide of aron. Hence I consider myself as entitled to conclude with the greatest certainty that neither cyanogen nor prussic acid as such are elements of anthrazothionic acid. And this conclusion is not only confirmed in the fullest manner by the decomposition of the acid by means of the galvanic battery; but also by the decompositions of the various combinations of anthrazothionic acid by means of heat to be related below. As Porrett affirms the contrary of this, there can be no doubt that he must have operated upon a salt containing a prussiate mixed with it. This would be more readily the case as he employed no alcohol, and indeed no method whatever to free anthrazothionate of potash from the prussiates.

Sect. 8 There are much stronger grounds for concluding from what has been already stated that ammonia is one of the immediate constituents of anthrazothionic acid; paradoxical as it may seem that an alkali should constitute an element of an

B

acid; for in all cases in which this acid was decomposed by means of sulphuric acid, or chlorine, I have been able at last to detect the presence of ammonia by means of potashley. Strong white clouds made their appearance, when a feather dipped in acetic acid was held over the liquid, and the ammonia at last became sensible by its smell. We might, perhaps, conclude, that when the anthrazothionic acid was decomposed, the water of the acid employed to decompose it (chlorine or sulphuric acid) supplied the hydrogen for the formation of the ammonia by uniting with the azote of the anthrazothionic acid, while the oxygen of the water might unite with the carbon of that acid and form carbonic acid; but in that case common muriatic acid must be capable of producing the same decomposition. Now as this is not the case, I conclude that the hydrogen which goes to the formation of ammonia during the decomposition cannot have been furnished by the water; but must undoubtedly have existed in the anthrazothionic acid itself together with the axote in the very proportion adapted for the formation of ammonia.

Sect. 9.-If a concentrated solution of anthrazothionate of potash in water be exposed to the action of a good Voltaic battery, a great evolution of gas takes place at the negative pole. This gas has a peculiar smell, similar to that of the inflammable gas from marshes. It is itself combustible, and when burned, carbonic acid gas is formed, and there remains a residuum of azotic gas. I consider this gas as a triple compound of carbon, hydrogen, and azote; though indeed it may be only a mixture of carburetted hydrogen and azotic gas. At the positive pole no gas is extricated; the liquid from the sulphur contained in it becomes yellowish, and at last allows the greater part of the sulphur to precipitate in large flocks. If silver or any other easily sulphuretted metal be placed in contact with the positive pole, it becomes immediately black by entering into combination with the sulphur; at the negative pole it remains completely white. It is possible that it may form a hydrate there, though the supposition is not very probable. The liquid, after having been exposed for some hours to the action of a battery of 100 pair of round plates, nine inches in diameter, was tried at both poles to discover in it the presence of prussic acid, nitric acid, and ammonia; but not a trace of one of these bodies could be found. The liquid from both poles, when mixed with a solution of iron in an acid, struck a blood-red colour, which must be ascribed to the presence of undecomposed anthrazothionate of potash. The sulphur at the beginning of the process remained in solution at the positive pole, and gave the liquor a yellow colour. This shows us that anthrazothionic acid is capable of existing with two proportions of sulphur, namely, a minimum, and a maximum. In this last state it must form peculiar compounds with the bases, which deserve hereafter to be accurately examined. If, as Porrett supposes, the acid were a compound

1

of prussic acid (or cyanogen) and sulphur, the former of these constituents would have appeared in abundance at the negative pole; and the liquid from that pole when mixed with a solution of iron and an acid would have formed a great deal of prussian blue.

Sect. 10.-One part of dry anthrazothionate of potash being mixed with five parts of chlorate of potash was inflamed by friction and percussion. Concentrated sulphuric acid inflamed the mixture still more violently; and when the experiment was attempted in a glass tube, a dangerous detonation took place. The Copper anthrazothionhydrate, considered by Porrett as an anthrazothionate of copper, exhibits the same phenomena as Porrett has remarked. I was in hopes in this way to have been able to have collected the azote from this last compound, and to have determined its quantity in a glass tube over mercury; but the instant the sulphuric acid came in contact with a quarter of a grain of hydrate and 14 gr. of chlorate of potash, a violent detonation took place, the tube was broken to pieces in my hand, and I received some slight wounds from the fragments of the glass scattered about in all directions. On another occasion, the same mixture took fire while I was rubbing it slightly in an agate mortar.

Sect. 11. Metallic Anthrazothionhydrates.-I consider all those bulky insoluble precipitates which take place when a solution of anthrazothionate of potash is poured into a solution of an easily reducible metal, as compounds in which the metal exists in the metallic state united with the anthrazothionic acid deprived of its hydrogen, that is to say, with anthrazothion, and with the water formed by the union of the hydrogen of the acid with the oxygen of the oxide, constituting a metallic anthrazothionhydrate. The following observations exhibit the grounds of this opinion. 1. When this precipitate is heated in a glass tube, after having /been dried for a day in a temperature between 122° and 144°, it always lets go a notable quantity of water, which collects in the cool part of the tube in the state of drops. 2. This water corresponds (at least I have found this to be the case in the copper anthrazothionhydrate) to the sum of the oxygen in the oxide and of the hydrogen in the acid. 3. During the escape of the water, which takes place at a temperature considerably under a red heat, the colour of the substance distinctly alters and becomes dark at the same time there separates a peculiar gaseous body with a particular smell, which I, both from the analogy of cyanogen, and because it is absorbed by ammonia and then strikes a blood-red colour with a solution of iron, consider as anthrazothion.*

** I regret very much that I have not been able to bring forward more direct proofs of the existence of anthrazothion, as no method which I attempted succeeded with me in separating it from the metallic anthrazothion compounds and procuring it in a separate stafe. Porrett must, in his experiments also, have employed only a very small quantity of this substance. It is much easier to procure cyanogen in