at a red heat; but the muriate of lime through which the gas afterwards passed, did not increase sensibly in weight. Hence every thing leads us to consider the hydriodates that have been melted or dried, as converted into metallic iodurets. The hydriodate of lime made with hydriodic acid, may be dried in the air without being decomposed. On the contrary, what has been made with iodine and line becomes deep coloured as it is concentrated, though we evaporate at a very moderate heat. The reason is, that this last hydriodate holds in solution a certain quantity of iodate, and these two salts have the property of decomposing each other, when brought to a certain degree of concentration. The hydrogen of the hydriodic acid and the oxygen of the iodic acid form water, and the iodine which is thus disengaged is dissolved in the undecomposed portion of hydriodate, and gives it a reddish brown colour. The whole hydriodate is not destroyed, because there is only a small quantity of iodate present; and when calcined in close vessels, it is completely freed from colour.*

Hydriodate of ammonia results from the combination of equal volumes of ammoniacal gas and hydriodic gas. It is usually prepared by saturating the liquid acid with ammonia. It is nearly as volatile as the hydrochlorate of ammonia; but it is more soluble and more deliquescent. I have obtained it crystallized in cubes. When heated in close vessels only a very small portion of it is decomposed. What sublimes is greyish black. If it be sublimed in contact of air, a much greater proportion of it is decomposed, and it becomes more or less coloured. It may be deprived of its colour by adding a little ammonia, or by exposing it to dry air. In the last case, the iodine to which it owes its colour is gradually volatilized.

The hydriodate of magnesia, formed by uniting its constituents together, is deliquescent, and crystallizes with difficulty. When heated to redness in close vessels, the magnesia abandons the acid in the same way as it abandons the hydrochloric acid. Having heated together iodine, magnesia, and water, to ascertain whether hydriodate and iodate was formed, as happens with the other alkalies, I obtained a flocky compound, which exactly resembled well prepared kermes. The liquid which coovered it was scarcely coloured, and I ascertained in it the presence of hydriodate and iodate of magnesia, but in very small quantity. When evaporated, a flea-coloured matter is deposited on the sides of the vessel, quite similar to that of which I have spoken; and towards the end of the process, the liquid acquires a deep colour. This phenomenon is owing (as in the case of lime) to the mutual decompo

To evaporate or calcine the hydriodates without the contact of air, I put them into a retort, to the beak of which I fit a tube, which, after having received the form proper for collecting gases, rises at its extremity parallel to the descending branch, and assumes nearly the shape of the letter U. When the aqueous vapour has expelled all the air of the retort, I place the ascending branch vader a glass jar filled with hydrogenous azotic gas, above the level of the water.

sition of the hydriodate and iodate of magnesia when they reach a certain point of concentration, but it is much more marked with magnesia.

The flea-coloured matter is decomposed when put upon burning coals. Iodine is disengaged, and magnesia remains. Potash decomposes it likewise. When boiled in water, its colour is not changed, but the liquid is found to contain a little iodate and hydriodate. If the quantity of water be considerable, pure magnesia remains, and the water contains iodate and hydriodate.

From these facts it appears that the flea-coloured matter is an ioduret of magnesia, and that its existence in water depends upon the property which the iodate and hydriodate of magnesia have of mutually decomposing each other when concentrated to a certain point. When the water is in great quantity, no ioduret of magnesia appears; but it is deposited as the concentration advances.

This phenomenon does not take place with the iodates and hydriodates of potash and soda. It begins to show itself with those of strontian. It becomes more sensible with the iodate and hydriodate of lime, and is very conspicuous with those of magnesia. But this last alkaline basis has a weaker affinity than the others; and it is perhaps because the oxides of zinc and iron, &c. have a still weaker, that they do not condense a sufficient quantity of hydriodic and iodic acids to prevent them from acting on each other, and that in treating them with iodine no iodates and hydriodates are formed, though these salts may be obtained separately.

(To be continued.)

ARTICLE X.

ANALYSES OF Books.

An Attempt to establish a pure scientific System of Mineralogy by the Application of the Electro-Chemical Theory, and the Chemical Proportions. By J. Jacob Berzelius, M. Ď. F. R. S. Professor of Chemistry at Stockholm. Translated from the Swedish Original by John Black. 1814.

THE doctrine of chemical proportions, though but recently introduced into chemistry, has produced a great reform in the science, and has given birth to a degree of accuracy, both in experimenting and reasoning, which has already placed chemistry on a footing with the mathematical sciences. Nobody has contributed more to produce this reform than Professor Berzelius. He has made the most numerous and accurate analyses which we at present possess, and has pointed out several general conclusions, which serve greatly to facilitate this kind of investigation. His activity, which surpasses that of any other chemical experimenter of the

present day, has led him to apply the doctrine of chemical propor tions to all the different departments of the science. The object of the present little work is to show that minerals are all real chemical compounds, that every species consists of constituents combined according to the laws of chemical proportions, and that they are susceptible of an accurate chemical arrangement into classes, orders, genera, and species, according to the nature of the substances of which they are composed.

The work bears evident marks of the great abilities and extensive knowledge of the author, and must suggest many important and useful ideas to every mineralogist who will read it with sufficient attention; though perhaps a more deliberate consideration of the subject, and a more minute acquaintance with the details of mineralogy, might have led to a modification of some of the opinions which Professor Berzelius has advanced; for instance, he seems to rate the knowledge of the external characters of minerals very low, and to consider the whole of the science of mineralogy as confined to an acquaintance with the constituents of which every mineral is composed. But it is necessary to recollect that, before the chemical analysis of any mineral can be of importance to the science, or lead to any useful inferences, we must be sure that the specimen which we subject to analysis belongs really to the species which we suppose, and that it is quite pure and unmixed with any other mineral. Now this knowledge can only be acquired by an acquaintance with the external characters of minerals a branch of knowledge which must therefore precede all useful chemical analysis. Hence it must always serve as the basis of our mineralogical knowledge. In fact, the labours of the chemist, who applies his practical skill to minerals, can only be of utility when he takes care to make his experiments upon correct and pure specimens. If a chemist, for example, analyze a specimen of mica, and publish the result under the name of an analysis of talc; or if he give the name of stilbite to what in reality is mesotype, his labours, instead of being useful, must be injurious to mineralogy. Yet these mistakes have been committed by chemists of acknowledged skill. The same injurious effects arise from the analysis of impure specimens, as when a mixture of felspar and quartz, or of felspar and garnet, is analyzed under the name of felspar. Unless I am much mistaken, errors of this kind have been lately committed by some of the most accurate analysts of the present day.

The knowledge of the constituents of minerals is always interesting, and in many cases indispensable. Thus the art of mining is founded on the knowledge of the different metals which may be extracted from the different ores. But to conceive that the whole science of mineralogy consists in a knowledge of the constituents of minerals, and that every thing else is of no consequence, is what no person can possibly do who has taken the requisite pains to make himself acquainted with the science. The diamond was as accurately distinguished by its external properties, and was applied to as

many uses by our predecessors, who were ignorant that this mineral consists entirely of pure carbon, as it is by us, who are acquainted with that fact-a fact which must be admitted to be curious and important; but not to constitute every thing of any value with respect to the mineralogy of the diamond. Gypsum was well known by its properties, and was applied to all the purposes for which it is used at present, before Margraaf and Lavoisier ascertained it to be a compound of sulphuric acid and lime. A mineralogist may be very well acquainted with the characters of gypsum, capable of distinguishing it from all other minerals, and aware of the different uses to which it is applied, though he be ignorant of the constituents which enter into its composition.

We must not, therefore, confine the science of mineralogy to the mere knowledge of the constituents of minerals. It includes many other particulars of great importance, and has frequently got the start of chemical analysis in its conclusions. Thus calcareous spar and arragonite were considered by mineralogists as two distinct species, even when the most expert chemists were unable to discover any difference in their composition. When chemical analysis shall have arrived at a state of perfection, we inay expect to find it agree in every respect with the conclusions drawn from the external characters; but in its present imperfect state, such discrepancies cannot be avoided; and when they do occur, it is but reasonable to give the superiority to the deductions from the external characters, as less likely to mislead us than an imperfect chemical analysis.

The object of Berzelius in the present little work is to show that all mineral species are really chemical compounds, composed of ingredients combined in definite proportions, and capable of being classified into orders, genera, and species, according to their composition, just as may be done with the salts. Though numerous analyses of minerals exist, yet it must be confessed that these definite proportions, this chemical composition according to the atomic theory, can be perceived only in a small number of individuals; while the great body of the mineral kingdom seems to bid defiance to the application of the laws of chemistry. But this discordance Berzelius considers, and I believe with justice, as only apparent, and not real. He ascribes it to three causes: 1. The inaccuracy of experimenters. 2. The mechanical mixture of foreign bodies with the chemical compounds, in consequence of the situation in which they were when they became solid or crystallized. 3. The deposition of two different compounds in contact with each other, which gives occasion to a form different from that of either of the compounds. Thus arragonite owes its form to the deposition of an atom of carbonate of strontian in contact with carbonate of lime at the time of its crystallization.

The minerals which it has hitherto been impossible to bring under the laws of chemical combination are the stones, composed chiefly of silica, alumina, lime, and oxide of iron, united in various proportions. Professor Berzelius conceives that in these

minerals the silica performs the function of an acid, and that it is chemically combined with the other earths or oxides which perform the function of bases. According to this notion, which has likewise been advanced by Mr. Smithson, the stony bodies are in reality salts. But they are often of a more complicated nature than the salts composed of the common acids and bases; for in the siliciates, as Berzelius terms these stones, we find not only the combination of silica with one base, but with two, three, or four bases, and often in various proportions, so as to constitute subsiliciates or supersiliciates.

If we suppose silica, alumina, magnesia, and lime, to be composed of one atom of oxygen and one atom of base, as I have done in the table published in the second volume of the Annals of Philosophy; or of two atoms of oxygen and one atom of base, as Berzelius has done, it is obvious that the number of integrant particles in any stony body may be determined by ascertaining the proportion which the oxygen of the various constituents bear to each other. Thus suppose we examine a mineral composed of silica and lime, and find that the oxygen in the silica is three times as great as in the lime, it follows that the mineral is composed of three integrant particles of silica and one integrant particle of lime; so that it may be termed a trisiliciate. This is the mode which Berzelius has taken to determine the constitution of the various stony bodies. It has the advantage of being at once easy and accurate, supposing us acquainted with the composition of the different earths.

Berzelius supposes silica to be composed of 50.36 silicon + 49.64 oxygen. If therefore we consider it as a protoxide, it follows that the weight of an atom of silicon is 1.007, and the weight of an integrant particle of silica 2007. He considers alumina as composed of 53.3 aluminum + 467 oxygen. The following table exhibits the weight of an atom of the different bases according to the analyses employed by Berzelius in this work.

This table will enable the reader to calculate the composition of the different stony bodies, which have been subjected to an accurate analysis. For example, schaalstein, or talle spar, is a hydrous bisiliciate of lime; somnite, or nepheline, is a siliciate of alumina; one of the species of calamine analyzed by Smithson, and composed of silica and oxide of zinc, is a siliciate of xinc; cerite is a siliciate of cerium. Berzelius gives examples of more complex VOL. V. N° IV.

U