Page images

series of products which includes the aluminates of zinc, iron, &c., together with secondary products.

C. H. B.

Artificial Production of Magnetite. By A. GORGEU (Compt. rend., 104, 1174-1177).-When a mixture of ferric and ferrous sulphates is decomposed in a bath of fused sodium sulphate, crystals of ferric oxide only are formed, and no magnetite is obtained. The latter is only produced if some reducing agent is introduced at the moment when the whole of the iron has been oxidised to the ferric state. Ferric oxide answers quite as well as the mixed sulphates.

When iron wire or turnings is introduced into fused sodium sulphate, a small quantity of sulphurous anhydride is evolved, and some free soda is formed, but the main product is a magnetic ferrous ferrite, containing a much larger proportion of ferrous oxide than magnetite. If the action of heat is continued, the ferrite undergoes oxidation, the sulphite and sulphide are reoxidised to sulphate, and eventually the whole of the iron is converted into the crystalline magnetic oxide. If the substances taken are weighed, it is found that the increase of weight at the close of the experiment is sensibly equal to the oxygen absorbed by the iron, the sodium sulphate having absorbed from the air a quantity of oxygen equal to that which it gave up to the iron. Potassium sulphate gives similar


Iron sulphides act energetically on fused alkaline sulphates, with evolution of a large quantity of sulphurous anhydride, and ultimate formation of magnetite and an alkaline sulphate. The alteration in weight is due to the loss of the sulphur originally present in the sulphide, and the absorption of 1:33 times its weight of oxygen by the iron in the sulphide. The yield is better the greater the intermediate formation of a double alkaline ferrous sulphide, and is best of all when a mixture of sodium sulphide and sulphite is used instead of the sulphate.

Artificial magnetite very closely resembles the natural mineral. It crystallises in opaque, magnetic octahedra with a metallic lustre, which are sometimes modified by very small faces of the rhombic dodecahedron; hardness 6·0—6·5; sp. gr. 5.21-5.25. The hardness and sp. gr. of the natural mineral are 5.5-6.5 and 4.9-5.27 respectively. The artificial crystals are not affected by water and carbonic anhydride at a bright red heat, are not attacked by nitric or hydrochloric acid diluted with 10 vols. of water, and are only very slowly dissolved by the strong acids or aqua regia. When roasted in the air, they yield pure ferric oxide, and the residue is usually not magnetic, but if the magnetite has been prepared from native ferrous carbonate the residue is magnetic, a result which is doubtless due to the presence of foreign ferrites. C. H. B.

Australian Bat Guano and Minerals occurring therein. By R. W. E. MACIVOR (Chem.. News, 55, 215-216).—The author gives the results of his examination of bat guano obtained from caves situated in various parts of Victoria. The only important deposit is at the Skipton caves, about 30 miles from Ballarat. Disseminated

through these deposits are crystals of Struvite (I), and of Hannayite (II), and Newberyite (III), minerals previously described by the author, of which the following new analyses are given :


MgO. FeO. MnO. (NH4)2O. P2O5.
I.... 16.07 0.81 0.16
II.... 18.54 0.31 0.09
III... 22.37 0.85 0.21

10.57 28.82
8.10 44.71


40.73 (35.84)

Reference is made to two new ammonio-magnesium phosphates, Muellerite and Dittmarite, which will be described later.

A. J. G. Destinezite. By G. CESARO (Jahrb. f. Min., 1887, 1, Ref., 412413). The author has made a chemical and crystallographical investigation of an almost white destinezite from Visé. On heating from 130° to 250°, 1 gram of the mineral lost 0·09 to 0.242 gram of volatile matter. At a red heat it lost 0·445 gram. Analysis gave the following results:


Hygroresidue. Fe2O3. P205. SO3. H2O. scopic H2O. Total. 1:40 37.60 16.76 18.85 25.35 0:30 100.26 From these results, the author calculates the formula P2O5,Fe2O3 + Fe2O3,2SO3 + 12H2O. He concludes that the constitution of the substance is O[Fe(OH)(SO,H) PO.]2 + 10H20. Under the microscope, it is seen that destinezite forms small crystals at least 0.01 mm. in length. They are colourless, and isomorphous with gypsum. B. H. B.

Wollastonite from Sardinia. By L. BUSATTI (Jahrb. f. Min., 1887, 1, Ref., 420).-The mineral examined was found at S. Vito, in the mining district of Sarrabus in Sardinia. It forms rosette-shaped aggregates of greyish-white fibres on a black Silurian clay slate. The substance is fusible with difficulty before the blowpipe, and gelatinises when heated with concentrated hydrochloric acid. Its hardness is 4, and its sp. gr. 2.7 to 2.8. From the analysis, the formula CaSiO, was deduced. The analytical results were as follows:


CaO. MgO.

Fe2O3. H2O. 49.78 45.12 1.20 2.20 0.60



B. H. B Griqualandite. By G. HEPBURN (Chem. News, 55, 240).-This mineral, from Griqualand West, South Africa, appears to be a pseudomorph after crocidolite, which it exactly resembles in structure. It has, however, the composition of a ferric silicate, 6SiO2,4Fe2O3,5H2O. It occurs as opaque, asbestos-like fibres, nonelastic, and of a snuff or golden-brown colour. The analytical results



SiOg. Fe2O3. combined. FeO. MgO. Moisture.
56.75 37.64 4.96 1.09 0.10



Total. 100.81

R. R.

3 b

The Meteorite of Karang-Modjo or Magetan in Java. By J. BossCHA (Jahrb. f. Min., Beilage 5, 126-144).-The Leyden Museum possesses a fine meteorite, described in the catalogue as the Magetan meteorite. It fell on October 3, 1883, at Karang-Modjo, in the Magetan district of Java. On the same day, the Ngawi meteorite fell in the same settlement. It is consequently highly probable that both meteorites belong to the same fall. The Magetan meteorite weighed 11911 grams. Its sp. gr. was found to be 3:34. In macroscopic and microscopic structure the Magetan and Ngawi meteorites are identical. B. H. B.

Sulphuretted Waters of Olette (Pyrénées Orientales). By E. WILLM (Compt. rend., 104, 1178-1180).—These springs are the hottest and most copious in the Pyrenees. There are also two which may be regarded as degenerated sulphuretted waters, since their general composition is the same, but they contain a larger proportion of sulphates, and no sulphides or thiosulphates. They also contain a notable quantity of nitrates.

[blocks in formation]





0.0602 0.0606 0.0592 0.0592 0.0606 0.0590 0.0046 0.0062 0.0051 0.0051 0.0037 0.0024 0.0020 0.0020 0.0024 0.0040 0.0092 Magnesium traces 0.0001 traces traces 0.0002 0.0003 Total carbonic anhydride 0.0410 0.0448 0.0444 0.0480 0.0520 0.0759

All the waters contain traces of phosphoric and boric acids, iodine, and arsenic, but are free from ammonia. The nitrates were estimated by converting them into ammonia by means of the zinc-copper couple.

The nitric acid is probably due to the nitrification of air which has come in contact with the water at a great depth, the change being assisted by the high temperature and pressure. The increase in the proportion of the calcium carbonate may be attributed to the action of the carbonic anhydride of this air on the surrounding rocks. similarity between the composition of the sulphuretted waters and of the degenerated waters shows that there has not been any infiltration of surface water. C. H. B.


Organic Chemistry.

Determination of the Relative Values of the Four Units of Chemical Activity of the Atom of Carbon. By L. HENRY (Compt. rend., 102, 1106—1109).—Hitherto there has been no actual determination of the relative values of the four units of chemical activity of the atom of carbon. Monocarbon-compounds of the type CX, will of course exist only in one form whatever the relative values of the four units, but if these values are different, monocarbon-compounds which contain two radicles should show differences. To take the simplest case, if one combining unit differs in value from the three others, compounds of the type CX,X' would exist in two forms, those of the type CX,X'X" in three forms, and those of the type CXX'X"X" in four. At present at least, the author's attention is confined to compounds of the type CX,X'. The best known and most easily obtained of these are the mono-substitutionderivatives of methane. It is true that only one variety of each of these derivatives is at present known, but this fact is without value, since it is impossible to affirm that in different specimens formed under different conditions the radicles saturate different combining units.

It is essential to prepare the mono-substitution-derivatives MeX by a systematic series of reactions, such that it can be affirmed that the radicle X is united successively with each of the four units of activity of the carbon-atom. In other words, it is necessary to ascertain if the order of introduction of the radicle into the molecule of methane exerts any influence on the properties of the resulting compound. The compounds selected for this purpose are the nitromethanes and the acetonitriles, which are well known and can be prepared by means of reactions that do not involve the use of high temperatures. The author uses the letters a, ẞ, y, to denote the compounds in which the first, second, third, and fourth atoms of hydrogen respectively are replaced by the radicle NO2 or CN.

The a-derivatives are obtained by the action of methyl iodide on potassium cyanide and silver nitrite respectively.

The B-nitro-derivative is prepared by the action of potassium nitrite on a salt of B-chloracetic acid, which is obtained by the action of chlorine on acetic acid formed from a-acetonitrile. The B-acetronitrile is prepared by the dry distillation of B-cyanacetic acid which is obtained by the action of potassium cyanide on ẞ-chloracetic acid.

The 7-derivative is prepared by similar reactions from y-chloracetic acid. This compound is formed by the dry distillation of 7-chloromalonic acid, which is obtained in the form of an ethyl salt by the action of chlorine on diethyl malonate, the malonic acid having been obtained from cyanacetic acid, CN, CH2COOH..

The 8-derivatives are formed in a similar manner from 8-monochloracetic acid, which is obtained by the action of heat on d-chloromethine-carboxylic acid, CCI (COOH), and this is prepared by the action of ethyl chlorocarbonate on ethyl monosodiomalonate.

The a-, B-, and y-derivatives have already been prepared by these methods and their properties investigated, but no differences could be detected. The c-derivatives are being prepared.

The whole argument is of course based on the principle of substitution and the stability of complex molecular structures throughout the course of a chemical reaction. C. H. B.

Action of Bromine on Isobutylene. By L. M. NORTON and H. J. WILLIAMS (Amer. Chem. J., 9, 87-89).-During the preparation of isobutylene bromide from pure isobutylene, it was noticed that a considerable quantity of a tribromisobutane was formed, boiling at 173-183° at a pressure of 235 mm. When treated with alcoholic soda, it yielded a liquid boiling at 158-161°; probably an isobutylene bromhydrin.

For purposes of comparison, a known tribromisobutane was prepared, namely, isobutylene bromide was converted by soda into isocrotyl bromide (b. p. 91-92°), and this saturated with bromine; the tribromide thus obtained boiled however at 155-161° at a pressure of 235 mm., and when treated with soda yielded a dibromisobutylene boiling constantly at 154-155°, and by addition of bromine it yielded a crystalline tetrabromisobutane, melting at 205°.

The two tribromisobutylenes are therefore not identical; that derived from isocrotyl bromide is CMe,Br: CHBr, and the only possible formula for that from isobutylene is CMe Br(CH,Br)2. This tribromide is not formed by the direct action of bromine on isobutylene bromide.

H. B.

Synthetic Acetonitrile. By L. HENRY (Compt. rend., 104, 1181 -1184).-Acetonitrile is obtained only in very small quantity by the action of methyl hydrogen sulphate on potassium cyanide. Much better results are obtained by the action of methyl iodide on potassium cyanide. Schlagdenhauffen has stated (Compt. rend., 48, 228) that these compounds will not react under ordinary pressure but only in sealed tubes at about 100°. This statement is not strictly correct. Methyl iodide, alone or dissolved in acetonitrile, has no action on potassium cyanide even at its boiling point, but if it be mixed with aqueous methyl or ethyl alcohol the reaction begins after a short time at the ordinary temperature, becomes more rapid, with continual development of heat, and the liquid soon enters into energetic ebullition. Only a very minute quantity of hydrocyanic acid, if any, is evolved, and the liquid remains colourless. The reaction is complete, and the yield is almost quantitative. The facility with which potassium cyanide is attacked by methyl iodide contrasts with the difficulty with which it is decomposed by ethyl iodide, and the simplicity of the chemical change is in contrast with the complexity of the corresponding reaction with the iodo-derivatives of the higher homologues.

« PreviousContinue »