free alkali, and a destruction of manganate and permanganate. These facts also explain the somewhat frequent failures in the preparation of the manganate. If the temperature is too high, reaction takes place between the manganate and the excess of dioxide.

Barium permanganate decomposes in a similar manner at 320°. If decomposition takes place in a closed vessel, the product is a dimanganite, 2MnO2, BaO, but if no precautions are taken to prevent access of moist air, the product is a heptamanganite, 7MnO2, BaO. Strontium and calcium permanganates behave similarly.

According to unpublished experiments of G. Lallement, metallic permanganates are converted into manganites at 100-150°. It is probable, therefore, that the method described is a general method for the preparation of metallic manganites. С. Н. В.

Non-existence of Chromium Heptasulphide. By G. BENDER (Ber., 20, 726-728).-The author attempted to prepare chromium. heptasulphide by the method employed by Phipson (Chem. News, 4, 125), and also by varying the conditions as much as possible, but obtained a mixed product containing less than half the amount of sulphur which would be contained in a heptasulphide.

N. H. M.

Lower Oxides of Molybdenum. By W. MUTHMANN (Annalen, 238, 108-137).-The preparation of molybdenum dioxide has been described by Svanberg and Struve (J. pr. Chem., 44, 257) and by Ullik (Annalen, 144, 227). It is conveniently obtained by fusing a mixture of anhydrous ammonium molybdate (8 grams), molybdic acid (7 grams), calcined potassium carbonate (14 grams), and boric acid (7 grams).

Molybdenum dioxide molybdate, Mo,O12, first described by Berlin, is prepared by heating a mixture of ammonium molybdate (1 part) and molybdic acid (2 parts). The product is repeatedly extracted with ammonia, and finally with strong hydrochloric acid, to remove a compound of molybdenum and nitrogen. This oxide is not attacked by alkalis, but it dissolves in warm strong sulphuric acid, forming a green solution, which easily parts with sulphurous anhydride, and turns blue. Nitric acid, aqua regia, and chlorine-water convert MosО12 into the trioxide.

The blue oxide of molybdenum, which is formed by the action of reducing agents on solutions of molybdic acid, or by boiling molybdenum-sulphuric acid, MoO3, SO3, with metallic molybdenum, has the composition Mo3Os. The olive-green oxide described by Berzelius appears to be a mixture of Mo2O, and the hydroxide Mo(OH)..

When molybdenum is dissolved in strong sulphuric acid, the dioxide is first formed, which oxidises to molybdic acid. The molybdic acid combines with the dioxide, forming Mo,O12. At a higher temperature sulphur dioxide is given off, and Mo3O, and MoO, are formed.

8

A dark-grey powder of the composition Zn,Mo,30 is formed by fusing sodium trimolybdate with metallic zinc. A similar magnesium compound has also been prepared. W. C. W. 20

VOL. LII.

Chlorostannic (Hydrochlorostannic) Acid. By K. SEUBERT (Ber., 20, 793-794).-Engel has described (Abstr., 1886, 984) a compound, H2SnCl + 6H2O, under the above name. The author, in conjunction with Schürmann, has also independently obtained the same compound. He states that the best and easiest means of preparing it is as follows:-Stannic chloride is mixed with concentrated pure hydrochloric acid in such proportions that the water in the acid shall be in relation to the stannic chloride as SnCl, : 6H2O. 100 parts of stannic chloride require 62 15 parts of 33 per cent. hydrochloric acid. Combination ensues, the mass becomes hot and softens, and hydrogen chloride is evolved. About 8 parts more of dry hydrogen chloride are then slowly passed into the mixture, which is then cooled. The whole solidifies to a scaly crystalline mass of the pure acid. The author gives the melting point of the pure substance as 19.2°. L. T. T.

Bromostannic (Hydrobromostannic) Acid. By K. SEUBERT and SCHÜRMANN (Ber., 794-796). The authors have obtained this compound in a way analogous to that employed (see preceding Abstract) for the preparation of the corresponding chloro-compound. 100 parts of melted stannic bromide were mixed with 741 parts of 50 per cent. hydrobromic acid. These proportions contain the necessary constituents for the formation of the compound H2SnBr, + 9H2O. The two liquids gradually mix, assume an amber colour, and the liquid gradually solidifies, leaving but very little mother-liquor. The pure substance crystallises in amber-coloured needles or triclinic plates, and melts at 47°. It is very deliquescent, and evolves hydrogen bromide when exposed to the air, making its analysis very difficult. The analyses agree most closely with the formula H2SnBre + 7H2O, but from analogy with the corresponding platino-compound the authors are inclined to believe the composition to be more probably H2SnBr. + 9H2O. The sodium salt, Na2SnBr、 + 6H2O, obtained by adding sodium carbonate to a solution of the acid, forms transparent, yellow needles. It is easily soluble in water, effloresces in the air, and at 90° loses all its water and stannic bromide.

L. T. T.

Gold. Part II. By G. KRüss (Annalen, 238, 30-77; compare this vol., p. 340).-The author reviews the various determinations of the atomic weight of gold which have been made from time to time by different chemists. For the purposes of this research, the purest refined gold of commerce was dissolved in aqua regia, evaporated to dryness with hydrochloric acid, and the residue redissolved and diluted with a large volume of water to precipitate the silver chloride dissolved by the acid liquid. The solution of auric chloride was further purified by one of the following methods:-I. The metal is precipitated by sulphurous acid, washed with hydrochloric acid and water, dried at 180°, and digested with strong sulphuric acid in a platinum dish to remove traces of silver. The residue is washed in hot water, dried, and fused in a platinum dish with potassium hydrogen sulphate to remove palladium, and afterwards fused with potassium

nitrate to remove iridium. Finally the gold is redissolved in aqua regia and reprecipitated by sulphurous acid. In methods II and III the gold is precipitated from the dilute solution of auric chloride by oxalic acid or ferrous chloride instead of sulphurous acid. IV. The gold is precipitated by ferrous chloride and purified as in method I; the metal is dissolved in aqua regia and reprecipitated by sulphurous acid. It is again dissolved, and finally precipitated by means of oxalic acid.

The author has mapped out the spark spectrum of gold; his results agree fairly well with those of L. de Boisbaudran (Spectres Lumineux, Paris, 1874), with the exception that six lines, of wave-lengths 560 1, 522-8, 521, 444-2, 434.5, 406-2, are missing. These lines observed by Boisbaudran are due to the presence of traces of platinum, palladium, and nitrogen. The temperature of the prism exerts a considerable influence on the position which the bright bands of the spectra occupy. Gold can be quantitatively separated from ruthenium, rhodium, and iridium by means of ferrous chloride, oxalic, or sulphurous acid. The separation of gold from platinum is best effected by oxalic acid, and from palladium by sulphurous acid.

W. C. W.

Mineralogical Chemistry.

New Zealand Graphite. By R. W. E. McIvor (Chem. News, 55, 125).-Two samples of graphite from Pakawan Bay, in Golden Bay, New Zealand, when dried, contained per cent.

Carbon
Ash..

...

34.99
65.01

51.45
48.55

D. A. L.

Natural Solutions of Cinnabar, Gold, and Associated Sulphides. By G. F. BECKER (Amer. J. Sci., 33, 199-210).-In the course of investigations on the geology of the quicksilver deposits of California, the author has taken up the question of the state of combination in which mercury is dissolved in natural waters. His experiments show that there is a series of mercury compounds of the form HgS, Na2S, one or the other of which is soluble in aqueous solutions of sodium hydroxide, hydrosulphide, or sulphide, and apparently also in pure water at various temperatures. These solutions subsist in the presence of sodium carbonates, borates, and chlorides. The waters of Steamboat Springs and of Sulphur Bank contain mercury in this form. Iron disulphide, gold, and zinc blende readily dissolve in sodium sulphide, and form double sulphides with sodium analogous to those of mercury. Copper also forms a soluble double sulphide, but combines more readily with sodium hydrosulphide than with the simple sulphide. All these soluble sulpho-salts may exist in the presence of sodium carbonate. Mercuric sulphide is readily precipitated from these solutions, diminishing temperature and pressure

being causes of precipitation. There are other natural methods of precipitation. Thus, mercury is precipitated by strong solutions of borax or hydrogen sulphide, or any stronger acid. At Steamboat Springs and Sulphur Bank, sulphuric acid is formed near the surface, and, percolating downward, precipitates mercury. Other ores and gold must be precipitated by the same causes. Another method by which mercury sulphide may be precipitated is mere dilution. This precipitation is accompanied by the formation of thiosulphate, a salt actually occurring in the waters of Steamboat Springs. The conditions of the solution and precipitation of ores described by the author are undoubtedly mainly instrumental in forming the deposits of Steamboat Springs and Sulphur Bank. Most of the other Californian quicksilver mines show similar ores and gangue minerals, and many of them are accompanied by warm springs, containing much the same salts in solution. Some of the gold deposits bear so close a resemblance to these deposits as to indicate that they also were formed by precipitation from solutions of soluble double sulphides.

B. H. B. Artificial Formation of Rubies. By FREMY (Compt. rend., 104, 737-738).-The methods employed by the author and Feil in 1877 for the artificial preparation of rubies were— -(1) To heat to bright redness in a fireclay crucible a mixture of alumina and red lead, with a minute quantity of potassium dichromate; and (2) to heat to a high temperature equal weights of alumina and barium fluoride with traces of potassium dichromate. The crystals were remarkable for their distinct shape, but were frequently lamellar. C. H. B.

Action of Fluorides on Alumina. By FREMY and VERNEUIL (Compt. rend., 102, 738-740).—The silica of the clay crucibles formerly used for the artificial preparation of rubies had no influence on the result, since the same effect is obtained in platinum or aluminium crucibles.

Almost all fluorides bring about the crystallisation of alumina at a red heat, but the experiments have been mainly confined to barium fluoride, calcium fluoride, and cryolite. When calcium fluoride and alumina are heated together in a platinum crucible at a very high temperature, a relatively small quantity of the fluoride exerts a very great mineralising power on the alumina. A mixture of 12 parts of alumina with 1 part of calcium fluoride becomes completely crystalline.

At the bottom of a platinum crucible, a layer of very pure transparent white fluorspar was placed, and on this a sheet of platinum pierced with minute holes. A thick layer of alumina, obtained by the calcination of ammonia alum and previously mixed with a minute quantity of chromic anhydride, was then added, and the crucible and its contents were heated to bright redness for several hours in a clay crucible, brasqued with alumina. After cooling, it was found that the calcium fluoride had fused, and the whole of the alumina was converted into crystals remarkable for their rose colour and their distinct form. Many fluorides, when heated in contact with moisture, evolve hydrogen fluoride, which can effect the crystallisation of amorphous

substances, like alumina.

It is, therefore, reasonable to suppose that hydrofluoric acid at a high temperature has played an important part as a mineralising agent.

The crystals obtained in this way are too small to have any commercial value.

C. H. B.

Zinc Ferrite: Artificial Production of Franklinite. By A. GORGEU (Compt. rend., 104, 580-583).-A solution of sodium sulphate (1 mol.), of zinc sulphate (1-2 mols.), and of ferric sulphate (0.25-0.5 mol.), was evaporated to dryness, and fused at a cherryred heat, a small quantity of the mixture being removed from time to time and treated with boiling water. When nothing but octahedra were seen in the liquid, together with some basic zinc sulphate, fusion was stopped, and the cooled mass was extracted with boiling water, the basic zinc sulphate being removed by means of dilute acetic acid. The formation of zinc ferrite under these conditions is due to the action of the ferric oxide, formed by the decomposition of the ferric sulphate, on the mixture of sodium sulphate and basic zinc sulphate. The ferrite can in fact also be obtained by the action of powdered hematite on a mixture of zinc and sodium sulphates, fused at a cherryred heat. If the hæmatite contains sand, crystals of willemite appear after the formation of franklinite, and before the formation of crystallised zinc oxide. This fact assists in explaining the association of these three minerals in certain veins.

Franklinite can also be formed from zinc chloride or fluoride. The chloride is heated in moist air with ferric chloride, or with powdered hæmatite. In the case of the fluoride, 4 parts are mixed with 6 parts of potassium fluoride, and 2 parts of ferric fluoride or 1 part of ferric oxide. Any crystals of zincite which are formed are removed by treatment with dilute acetic acid.

Franklinite obtained by these methods crystallises in regular octahedra, modified by facets of the rhomboidal dodecahedron. Small crystals are transparent, have a reddish-brown colour, and are monorefractive; larger crystals are opaque, and have a metallic lustre. They are not magnetic, are not affected by heat, and are only very slowly attacked by acids; hardness, 65; sp. gr. 5:33.

Natural crystals of franklinite are opaque, and give a brownishblack streak, whilst the streak of the artificial product is yellowishred. Moreover the natural mineral is magnetic, and its sp. gr. does not exceed 5'09. These differences are due to impurities in the natural mineral. A franklinite identical with the natural mineral is obtained by fusing together sodium sulphate and 10 per cent. each of zinc, manganese, and ferric sulphates, and adding a small quantity of a reducing agent, such as ferrous sulphide, during fusion. The small quantity of ferrous oxide thus formed combines with some of the ferric oxide and forms magnetite, which crystallises with the zinc ferrite. C. H. B.

Plumbocalcite from Wanlock Head. By A. LACROIX (Jahrb. f. Min., 1887, Ref., 1, 238—239).---Five varieties of plumbocalcite from Wanlock Head, Scotland, had the following composition :