10 ABSTRACTS OF CHEMICAL PAPERS. lnorganic Chemistry. Reactions of Hydrogen Peroxide. By ARNOLD NABL (Moncctsh., 1901, 22, 737-744. Compare Abstr., 1901, ii, 16, 94)-Hydrogen peroxide and sodium thiosulphate react according to the equation : 2Naj2S20, + H202 = 2NaOH + 2Na,S40,. If the alkali is not neutralised, 75 per cent. of the thiosulphate remains unoxidised and the reaction leads to the formation of sulphate and dithionate as well as tetrathion- ate : 2Na2S20, + 7H202 = 2Na2S04 + H2S206 + 6H20 j H2S206 + H202 = 2H,SO,. Barium sulphite and hydrogen peroxide give a small quantity of dithionate as well as of sulphate when the sulphite is in excess, The reactions represented by the equations 2H2S0, + H202 = 2H,O + H2S206 and H2S206 + H202 = 2H2S04, therefore take place simultaneously.K. J. P. 0. Alkali Salts of Hydrogen Peroxide in Aqueous Solution. By HARRY T. CALVERT (Zeit. physikal. Chem., 1901, 38, 513-542)- An historical account of the peroxides of the alkali metals is given. For the experiments, the hydrogen peroxide was prepared by repeated distillation until the conductivity was constant, and then concentrated on the water-bath. The distribution ratio of hydrogen peroxide between water and ether a t 20' is 15.6 and is independent of the con- centration. A constant ratio (7.03 a t 25*, and 6.65 a t 0') was also found when the ether was replaced by amyl alcohol, and this is not altered by the addition of acids, I n presence of alkalis, the distribu-INORGANIC CHEMISTRY. 11 tion ratio is increased and the curve representing the change of ratio with increasing concentration of hydrogen peroxide approaches asymptotically to a line denoting the ratio on the assumption that 1 mol. of alkali fixes 14 mols.of hydrogen peroxide. Addition of hydrogen peroxide diminishes the saponifying power of sodium hydroxide, indicating that hydroxyl ions disappear. The conductivity of hydrogen peroxide solutions was determined in a modified Kohlrausch cell, in which the electrodes consisted of tinned iron, which does not catalyse the solution. The conductivity of alkali salt solutions is very slightly diminished by addition of hydrogen peroxide ; that of solutions of hydroxides of the alkali metals is very greatly reduced, This is explained on the assumption t'hat with the -hydroxyl ions tke hydrogen peroxide forms superoxide ions, the migration-velocity of which is small compared with that of the hydroxyl ions.Using the Ostwald-Walden rule, the author calculates the migration-velocity of this new anion to be 48.5 (Kohlrausch and Holborn units), the same value being found from solutions containing the cations Li., Na*, K*, Rb*, and Cs*. The migration of the superoxide anion has been proved experi- mentally by the method described by Noyes and Blanchard (Abstr., 1901, ii, 91), lead oxide being used as indicator. From the depression of the freezing point of water containing sodium hydroxide and hydrogen peroxide, using excess of the latter to diminish the hydrolysis, it is shown that the anion is univalent and is derived from the liydroxyl ion and neutral hydrogen peroxide (Abegg and Bodlander, Abstr., 1899, ii, 542).The results are in agreement with the assumption that the ion is O', and the compound formed from sodium hydroxide and hydrogen peroxide is NaO,. The solubility of potassium chlorate in hydrogen peroxide is much greater than that in pure water, and consequently such a determination could not be used to ascertain if the hydrogen peroxide forms a com- plexion with the cation. Molecular Compounds of Hydrogen Peroxide with Salts, By SIMEON L. TANATAR (Zeit. anorg. Chem., 1901, 28, 255-257).- The compound KF,H,O, is obtained by dissolving potassium fluoride in 15 per cent. hydrogen peroxide and evaporating a t 5OOso long as no serious decomposition occurs. It crystallises in monoclinic needles, is not hygroscopic, but exceedingly soluble in water, is not decomposed at 70" and only partially so at l l O o , and is fairly stable when dry.A similar compound is obtained by dissolving sodium sulphate in 3 per cent. hydrogen peroxide and has the composition Na2S0,, 9H,0,H20,. With sodium nitrate, the double salt, NaNO,,Na,O,,SH,O, is obtained. It is very unstable. Generalisations on Halogeln Double Saits. By HORACE L. WELLS (Amer. Chem. J., 1901, 26, 389--408).-A long list is given of halogen double salts of the alkali metals, ammonium, and univalent thallium, with negative metals ; the salts are arranged according to types, which are designated by ratios indicating the number of atoms of each metal present. J. McC. E. C. R.12 ABSTRACTS OF CHEMICAL PAPERS. The remarkable similarity in the prominent types of the series of different valencies leads t o the conclusion that the valency of the metal of a negative haloid has no influence on the types of double salts which i t forms.The molecules of alkali haloids have nearly the same combining power as molecules of negative haloids. Salts of simple types (particularly the 2 : 1 and 1 : 1 ratios) predomi- nate. Remsen’s law which states that the number of alkali haloid molecules which can combine with a negative haloid molecule is not greater than the valency of the metal of the latter, must be aban- doned. The double haloids appear to increase in variety and ease of formation from the iodides t o the fluorides. They may be classified in three groups, based upon their behaviour in solution.(1) Salts, such as potassium platinichloride, which undergo ionisstion into alkali metal ions and complex negative ions. (2) Salts which readily separate into their component haloids in solution, but can be recryst,al- lised unchanged from water or from dilute acid solutions. (3) Salts which require the presence of an excess of one of their component haloids i n solution for their formation. E. G. By YUKICHI OSAKA (Zeit. yhysikal. Chem., 1901, 38, 743--749).--The addition of iodine to a solution of potassium iodide or hydrogen iodide produces a rise of the freezing point proportional t o the quantity of iodine added, and greater for the hydrogen than for the potassium salt. Hence it follows that the total concentration of ions and undissociated molecules is decreased by the addition of iodine. This necessitates a greater affinity constant for the iodides than for the tri-iodides, so that Damson’s assumption t h a t these affinity constants are equal is incorrect (Trans., 1901, 70, 238).Tri-iodides. L. M. J. Influence of the Concentration of the Hydrogen Ions on the Action of I o d a t e s on Haloid S a l t s . By HUGO DITZ and B. M. MARGOSGHES (Zeit. angew. Chem., 1901, 14, 1082-1091).- Potassium iodate and iodide readily react in the presence of a sniall amount of an acid (hydrogen ions) liberating iodine, and the amount thus deposited is directly proportional t o the amount of acid present (Fessel, Zeit. anorg. Chem., 1900, 23, 66). Potassium bromide and iodate do not react so readily in the presence of an acid (Bugarszky, Abstr., 1896, ii, 216) and the iodine ions are only transformed into free iodine when the concentration of the hydrogen ions exceeds a certain minimum.Potassium chloride reacts less readily than the bromide and the necessary concentration of hydrogen ions is much greater. When definite amounts of hydrochloric or sulphuric acid are zdded t o a potassium iodide solution mixed with an excess of iodate, the amounts of iodine liberated and of iodate left are found t o corre- spond with t8he amounts required for the given quantities of acid employed. The free iodine was extracted with toluene and titrated with N/lO thiosulphate and the residual iodate was titrated by means of the same reagent. When acetic acid is added to a potassium iodide-iodate mixture, the reaction is not normal and the residual iodat,e is always less than that required by theory j similar resultsINORGANIC CHEMISTRY.13 have been obtained when acetic acid was used i n presence of sodium acetate. The anomaly is probably due to the formation of an organic kiYl%Tl3 u i j va i?i g-6, Boric acid is not capable of liberating iodine from an iodide-iodate mixture except in the presence of glycerol or dextrose. The same acid does not liberate free halogen from a bromide-iodate mixture, wen in the presence of glycerol. Phenol also is incapable of liberat- ing iodine, but picric acid liberates a small amount from an iodide- iodate mixture. With mixtures of bromide, iodide, and iodate in the presence of acetic or hydrochloric acid, the amount of iodine liberated corresponds with the reaction between the iodide and iodate.A bromate-iodide mixture also yields iodine on treatment with acids, but requires the addition of several C.C. of N/lO acid before the liberation of iodide is started. A bromide-bromate mixture in the presence of acetic acid and. sodium acetate yields no free halogen. A chlorate-iodide solution, even in the presence of considerable! excess of dilute hydrochloric acid liberates but little iodine ; coneen-- trated acid, on the contrary, liberates a much larger amount. J. J. S. Supposed Anomalous Behaviour of Oxygen at Low Pres- sure. By MAX THIESEN (Ann. Phys., 1901, [iv], 6, 280-301).-Tht? author's observations are quite unfavourable t o the supposed existenc6: of an anomaly for oxygen under 0.7 mm.pressure (compare Bohr, Ann. Yhys. Chern., 1886, [ii], 27, 459; Eayleigh, Abstr., 1901, ii, 542). J. C. P. Dissociating Power of Hydrogen Sulphide. By WM. T. SKILLING (Afizer. Chem. J., 1901, 26, 383--384).-When a solution o;F potassium chloride in liquid hydrogen sulphide is placed in a tube pro-- vided with platinum electrodes and connected with a battery ojE 40 volts, no conduction takes place. E. G. &+a1~3+j,c: T?wu+jAwis. 11. Drnxt~~fisi+i~~~~ ~f ChlAwnwiL-- phonic Acid into Sulphuryl Chloride and Sulphuric Acid. By OTTO RUFF (Be?.., 1901, 34, 3509--3515).-At 170°, the re- action, 2S03HC1=S3,C1,+ H,SO,, isareversible one, and after 72 hours, the equilibrium reached is 1SO,Cl2 : 1H2S0, : 2.5S03HC1 ; sulphur di- oxide and chlorine are not formed between 170'and 190°, although a t a higher temperature they begin to be noticeable.Obviously, therefore, the sulphuryl chloride is not formed by the union of sulphur dioxide and chlorine, initially produced by complete dissociation of the chloro- sulphonic acid (compare Heumann and Kochlin, Abstr., 1883, 781), but is a direct decomposition product ; this is emphasised by the fact that, when the formation of the sulphuryl chloride occurs at the boil- ing point of the chloiosulphonic acid owing to the presence of a catalytic agent its amount is not increased by the passage of a current of chlorine and sulphur dioxide through the liquid. When sulphur dioxide and chlorine are produced at a high temperature, they14 ABSTRACTS OF CHEMICAL PAPERS. are due to the latter causing decomposition in the sense of the equation Mercuric salts rapidly decompose chlorosulphonic acid a t its boiling point into sulphuryl chloride and sulphuric acid, and several other salts effect a similar result, only more slowly.The figures after the names of the following salts indicate the number of grams of sulphuryl chloride formed by boiling 50 grams of the acid for 60 minutes with 1 gram of the salt. Nercuric chloride and sulphate, each 13.0 ; mer- cury, 12 ; antimony penta- and tri-chloride, 7.5 ; stannic chloride, 5.8; bismuth chloride, 3.3 ; platinic chloride, 2.5 ; uranyl chloride, 1.7 ; gold chloride, 1.2 ; copper sulphate, 0.8 ; tungsten chloride, 0.8 ; lead chloride, 0.7 ; cobalt sulphate, 0.5, and magnesium chloride, 0.5 ; the chlorides of zinc, aluminium, iron, calcium, and sodium are without action.As the whole of the chlorosulphonic acid is easily decomposed by boiling with mercuric chloride, the method is probably capable of commercial application for the manufacture of sulphuryl chloride, The mercuric chloride is not changed a t all in the action, but mer- curic sulphate is converted into mercuric chloride ; sodium chloride dissolves in chlorosulphonic acid with evolution of hydrogen chloride and production of the sodium salt, ONa*SO,*Cl, which is precipitated by the addition of sulphuryl chloride and is readily decomposed by water. W. A. D. SO, + 2HC1= SO, + H20 + Cl,. Formation of Dithionic Acid. By JULIUS MEYER (Ber., 1901, :34, 3606--3610)-The formation of barium dithionate, by passing r;ulphur dioxide into water in which manganese dioxide is suspended, i's probably to be represented by the following equations : 2Mn0, + :3H,SO, = Rln,(SO,), + 3H,O + 0 = MnS,O, + MnSO, + 3H,O + 0,asman- i;anous sulphite and sulphate are always formed in appreciable quan- t;ity.Ferric hydroxide and sulphurous acid, in the absence of air and c t t low temperatures, yield ferrous sulphite and ferrous dithionate, lFe,(S0,)3 = FeSO, + FeS,O,.. Cobaltic and nickelic hydroxides react iivith sulphurous acid in a similar manner. Lead, barium, magnesium, :md sodium peroxides do not yield dithionates on treatment with sul- ]?hurous acid (compare Carpenter, Trans., 1902, 81, 1) The electrical conductivities and freezing points of barium dithionate r;olutions are given, and the formula H,S,O, for the acid is considered lto be proved (compare Ostwald, Zed.physikal. Chem., 1887, 1, 106). J. J. S. So-called Sulphimide. By ARTHUR HANTZSCR and A. HOLL ((Be?., 1901, 34, 3430-3445. Compare W. Traube, Abstr., 1892, 1389; 1893, ii, 268)-Sulphamide is unimolecular, and, in the pure rstate, not an electrolyte, and corresponds with carbamide ; sulphimide, Ion the other hand, is termolecular and corresponds with cyanuric acid. lsulphamide is best purified by repeated crystallisntion from ethyl :alcohol ; it forms rhombic plates melting at 91.5' (Traube, 75-81°), .is perfectly tasteless, and quite neutral. I n the preparation of sulph- :amide by the action of sulphuryl chloride on a light petroleum solu- .tion of ammonia, several bye-products are formed, but the composition (of these has not been determined.INORGANIC CHEMISTRY.15 Trisulphimide, O H * S O < ~ ~ ~ $ ~ ~ { > N , is crystalline, but the yield \ , is extremely small, and the analytical results do not agree with those required for the pure compound; it may be, however, that it contains combined water. It crystallises from methyl alcohol in colourless, glistening needles, melts at about 161°, is odourless, but has a sharp, acid taste, dissolves in alcohol, and also sparingly in ether, but is insoluble in benzene or chloroform. Aqueous solutions are fairly stable, except when warmed. Ebullioscopic determinations of the molecular weight in ethyl acetate solution point to the formula (SO,*NH), The values obtained for the molecular conductivities of trisulphimide and its salts in aqueous solution show (1) that trisulph- imide behaves as a strong acid; (2) that it is tribasic.Further, the conductivity of the potassium salt is nearly the same as that of potassium ferricyanide. The molecular conductivity of silver trisulph- imide is much lower than that of potassium trisulphimide a t the same dilution. Pyridine silver trisulphimide, (S02N),Ag3,6C5NH5, crystallises in prisms and is gradually decomposed at 140". The N-methyl derivative, NMe*So2>NMe, obtained from the silver salt, forms colonrless SO,<NMe* so, crystals melting at 121' and readily soluble in most organic solvents. It is not hydrolysed by alkalis and only very slowly byacids, yielding met hylamine and sulphuric acid.Tribenxo y ZsuZphimide, (SO,N* COPh) 3, crystallises in prisms melting a t 112'. Isomorphism of Selenates and Tellurates. By JAME~ F. NORRIS and WILLIAM A. KINGMAN (Amer. Chem. J., 1901, 26, 318--324).--The authors have attempted to prepare isomorphous selenates and tellurates, but without success. All the tellurates except those of the alkali metals are insoluble and the soluble tellurates do not resemble the corresponding selenates in crystalline form, solubility, or amount of water of crystallisation. Rubidium hydrogen tellurute, RbHTeO,,&H,O, is a crystalline salt, soluble in about 20 parts of cold water, and slightly more so in hot water. The cesium salt, CsHTeO,,&H,O, forms small, cubic crystals and is soluble in about 30 parts of cold water.Rubidium hgdrogen selenate, RbHSeO,, is a hygroscopic, well crystal- lised salt, which dissolves in about its own weight of water. The cesium salt, CsHSeO,, crystallises in large, flat plates with pointed ends and is extremely hygroscopic. Rubidium tellurate, Rb,TeO,,SH,O, crystnllises in prisms with pyramidal ends and is soluble in about 10 parts of water. Aqueous Ammonia Solutions. By FRANZ GOLDSCHMIDT (Zeit. ccnorg. Chem., 1901, 28, 97--139).-The partial pressure and the conductivity of ammonia solutions (0*5N, 0*75N, and IN) were deter- mined and the same constants for solutions to which carbamide (1N and 1.5N) had been added. The vapour tension was measured by the method previously used by Gaus (Abstr., 1901, ii, 7). The increase of the partial pressure of the ammonia is almost, exactly proportional to the amount of carbamide added.The values of k found from the expression J. J. S. E. G.16 ABSTRACTS OF CHEMICAL PAPEItS (k + H20)/(k + H20’) = A2p’/AI2p is negative and almost exactly constant; k is the hydration constant, H,O the active mass of the water (taken as loo), H20’ the active mass after the addition of the carbamide, X and A’ are the conductivities, and p and p’ the ammonia partial pressures of the solutions with and without the carbamide. The variation of the active mass of the water is taken as proportional to the variation of the vapour pressure, on the assumption that only a monohydroxide is formed. KO significance can be attributed t o the fact that the value of k is negative. From the conductivity of ammonia solutions (which show a maximum molecular conductivity), it is found that the dissociation constant ( K ) varies from 22.1 x to 0.23 x 10-6 for solutions which are 0.0109N to 12.89 A‘.As the solution becomes dilute, the value of Kapproaches a constant, and assuming that this begins a t the concentration 0*02N, the value of K is 19.1 x 10-6. The inconstancy of the values of K cannot be attributed to the formation of complex ions such as NH,*NH,*, for the lowering of the freezing point of water containing ammonia and ammonium salts corresponds with the value calculated for the quantity of material added. The addition of ammonia diminishes the conductivity of aqueous salt solutions. The diminution is directly proportional to the quantity of ammonia added and amounts t o 2 to 3 per cent.of the value of the conductivity per gram-mol. of ammonia per litre. Lithium salts are most affected in this way by ammonia, sodium salts less so, and potassium salts least. The action of ammonia is independent of the nature of the anion present. It is concluded that the speed of migra- tion of the ions is reduced by the presence of ammonia, and this is also proved thermodynamically. The influence of ammonia on the conductivity of ammonium chloride, mono-, di-, and tri-methylamine hydrochloride and piperidine hydro- chloride is of the same nature as, but much smaller than, that on the alkali salts ; with tetramethylammonium chloride, the eRect is of the same magnitude as with potassium nitrate. J. McC. Chemical Equilibrium in the Reduction of Nitric Acid by means of Nitric Oxide.By A. V. SAPOSCIINIKOFF (J. Buss. Phys. Chen8. Xoc., 1901, 33, 506-516. Compare Abstr., 1900, ii, 722).- The author’s previous experiments ( h e . cit.) on the decomposition of nitrous acid according to the equation, 3HN0, = HNO, + 2N0 + H,O, having failed to yield the equilibrium constant of the reaction, he has now studied the inverse change, the method of experiment being to pass nitric oxide through nitric acid of a cert’ain strength and to determine from time to time the electrical conductivity of the solution. A t the end of each experiment, the amount of nitrous and nitric acids and of nitric oxide in the liquid was determined. On calculating the constants of equilibrium for acids of varying concentration from the formula, K=cY/x2c12, where c and c1 are the concentrations of the nitrous and nitric acids respectively in the final solution, and x the degree of dissociation of the nitric acid, it is found that for nit,ric acids of AT t o i V / l O initial concentration, the constant ( x 10,000) varies within the limits 142 and 178 and has a mean value of 159; forINORGANIC CHEMISTRY.17 acid of 0*05Nconcentration, the constant is 232, the high value being probably due to the fact that at such great dilution the reaction proceeds very slowly and possibly does not reach its final point; with acids of higher concentration than normal, the constant falls regularly, a behdviour probably explained by the final product of the reaction consistii-g partly of oxides of nitrogen mixed with the nitrous acid.The coefficient of absorption o€ nitric oxide in litres of gas per litre of acid is given for each dilferent nitric acid employed ; Kahlbaum’s acid, having H, sp. gr. 1.517 at 15’/4”, contains 0.88 per cent. of nitric oxide, the coefficient of absorption in this case being 12.5. The speed of the abmrption of nitric oxide by nitric acid varies to a very large extent with the concentration of the acid, as is shown by curves connecting litres of the gas absorbed with time in hours. By RUDOLF WEGSCHEIDER and FELIX KAUFLER (Monatsh., 190 1,22, 700-706).-Red and yellow phosphorus may be either polymorphous or chemically different (isomeric or poly- meric). If the two forms are polymorphous, the liquid forms must be identical, and molten yellow phosphorus or a concentrated solution should, on addition of red phosphorus (which is the stable form), change into the latter.Experiments show that such is not the case. A saturated solution of yellow phosphorus in carbon disulphide sown with red phosphorus shows no perceptible change, and, on lowering the temper- ature, yellow phosphorus separates. Similar molten yellow phosphorus at 20Oo, to which red phosphorus has been added, does not change, The two forms are not polymorphous, but chemically different. T. H. P. Allotropy of Phosphorus. K. 5 . P. 0. Metaphosphates. By ARTHUR WEISLER (Zeit. anos-g. Chsm., 1901, 28, 177--909).-Sodium trimetaphosphate, when prepared according to Flei t mann’s and Henneberg’s methods, and when prepared from sodium hydrogen phosphate and ammonium nitrate according t o Knorre’s method (Abstr., 1900, ii, 651), has an electrical conductivity which indicates that it is a salt of a tribasic acid.The trimetaphosphate, in aqueous solution (1/32N), is not altered by boiling ; neither orthophos- phste nor pyrophosAphate is formed. Bdrium, manganese, and silver trimetaphosphates are described ; of these the manganese salt has a n electrical conductivity corresponding with that required for a salt of a tribasic acid. The other two salts are too insoluble for the determination of their electrical conductivities. Cropper trimetaphosphate could not be obtained from sodium trimeta- phosphate, tbe product being a pyrophosphate of the formula Cu2P,07,5H,0. I n the case of zinc, a sodium zinc pyrophosphate is produced.The sodium cadmium trimetaphosphate, CdNa,(P0,)6,4 H20, is obtained by adding cadmium iodide to the sodium salt. Sodium hexametaphosphate, prepared from sodium pyrophosphate according t o Knorre’s method (Zeit. angew. Chenz., 1892, 641)’ has a n electric conductivity A,, = 31.2, which is evidence that the salt has a more complicated composition than the trimetaphosphate (A,? = SY.4). The hexametaphosphate is easily decomposed in aqueous solution and LXXXII. ii. A d when heated at 40” yields the pyrophosphate. E. c. K. >18 ABSTRACTS OF CHEMICAL PAPERS. Chemical Reactions produced by Radium. By MARCELLIN P. E. BERTHELOT (Compt. rend., 1901,133,659-664).-The radiations from radium decompose iodic acid, with liberation of iodine, and also decompose nitric acid, the changes in both cases being endothermic.They do not, however, promote the oxidation of oxalic acid, or the conversion of sulphur into the variety insoluble in carbon disulphide, nor do they, like the silent electric discharge, cause the polymerisation of acetylene. The author confirms the statement that the rays gradually turn glass black, a change which he attributes to the reduc- tion of lead compounds to the metallic state, and has also observed the production of a violet colour similar to that produced by the action of light on certain glasses containing manganese. It would seem, therefore, that the radiations cause a reduction and an oxidation simultaneously, and possibly the one change is consequent on the other. C.H. B. Action of Bromine on Metallic Silver in the Light and in Darkness. By V. VON CORDIER (Monatsh., 1901, 22, 70’7-716. Compare Abstr., 1900, ii, 343,723).-By use of a specially constructed apparatus, the action of bromine on metallic silver illuminated by an arc lamp, an incandescent gas lamp, or diffused daylight, and in dark- ness was investigated. Whilst light assists the combination of silver and chlorine (Zoc. cit.), it hinders that of silver and bromine, Bromine is not given off in the light from silver bromide in the presence of carbon dioxide. K. J. P. 0. Action of Hydrogen Peroxide on Silver Oxide. By MARCELLIN P. E. BERTHELOT (Compt. vend., 1901, 133, 555-569).-The author has investigated in the calorimeter the action of several acids on (1) silver oxide, and (2) silver oxide which had been treated with hydro- gen peroxide.The development of heat differs considerably in the two cases, both in rate and amount, and the results, which are described in detail, confirm the author’s previous conclusions as to the formation of a higher oxide of silver (Abstr., 1899, ii, 149). C. H. B. Reduction of Copper by Solutions of Ferrous Salts. By H. C. BIDDLE (Amer. Chern. J., 1901, 26, 377--382).-The precipita- tion of copper by solutions of ferrous salts is a reversible action, the direction of which in any case is determined by the relative concentra- tion of the ferrous, ferric, and copper (cuprous and cupric) ions. This statement is justified by the following experimental evidence, I n a solution containing an appreciable quantity of ferric ions, or in which these would be formed in the course of the reaction, copper is not deposited ; this is shown by the fact that ferrous chloride and sulphate are incapable of reducing the corresponding copper salts.In a solu- tion containing a few ferric ions, and in which the reaction does not cause an appreciable increase of them, a sufficient concentration of ferrous and copper ions will produce the deposition of copper. The tendency of ferrous salts to reduce those of copper is shown by the precipitation of cuprous thiocyanate when ammonium thiocyanate isINORGANIC CHEMISTRY. 19 added to a solution of ferrous and cupric chlorides. From a mixture of ferrous and cupric hydroxides, crystals of cuprous oxide slowly separate. When an excess of ammonium carbonate is added to a solution of ferrous and cupric chlorides, a yellow liquid is obtained which gradually deposits a mirror of metallic copper.If a solution of cupric and ferrous chlorides is treated with sodium carbonate in slight excess or with potassium hydrogen carbonate, reduction slowly takes place with loss of carbon dioxide and formation of basic ferric carbonate and copper. E. G. Mixed Crystals of Copper Sulphate and Zinc Sulphate. By H. W. FOOTE (Amey. Chern. J., 1901, 26, 418--428).-1t is well known that from solutions of copper sulphate containing zinc sulphate an isomorphous mixture of triclinic crystals separates consisting of CuS0,,5H20 together with a smaller quantity of ZnS0,,5H20. AS the quantity of zinc sulphate is increased in the solution, an iso- morphous mixture of monoclinic crystals is obtained containing CuS04,7H20 and ZnS04,7H20, and in presence of a still larger pro- portion of zinc sulphate, rhombic crystals of ZnS04,7H20 with a small amount of CuS0,,7H20 are produced.It has been shown by van't Hoff, from theoretical considerations, that the composition of mixed crystals a t their mixing limit' (which represents the com- position of mixed crystals of one salt with a maximum of another) ought to be a function of the temperature ; the same conclusion is arrived a t by application of the phase rule. The authors have carried out experiments on the mixing limit ' of copper sulphate and zinc sulphate a t 12', 25', 35'; 40°, and 4 5 O , which confirm the accuracy of the above assumption, and also show that whilst the composition of mixed crystals varies with the temperature, the salts mentioned do not form completely isomorphous crystals between 12' and 56'.It is also found that i n solutions yielding two forms of crystals, the amount of copper sulphate in solution remains nearly constant, whilst the quantity of zinc sulphate increases con- siderably with rising temperature. By MAX GRGCIER (Zeit. anorg. Chern., 2901, 28, 154--161).-When cuprous chloride is treated with water in an atmosphere of hydrogen or carbon dioxide, the chlorine passes almost completely into solution and a dark red residue, consisting of cuprous oxide and copper, is left. The separation of the copper is due to the action of light, for when the extraction is carried out in the dark the residue is almost pure cuprous oxide, Water acting on cuprous chloride in presence of a little air gives an orange-red residue of cuprous oxide mixed with about 6 per cent.of basic cupric chloride. The amount of cupric compound left in the residue is always small, showing that most, of the oxidation product of the cuprous chloride passes into solution, By the action of water and air sufficient for the complete oxidation, a residue was obtained which had the composition 3CuO,CuCl2,4H2O. The primary action is the hydrolysis of the cuprous chloride, and the liberated hydrochloric acid, in presence of oxygen, acts upon more cuprouB chloride forming the cupric corn- E. G, Cuprous Chloride. 2-220 ABSTRACTS OF CHEMICAL PAPERS, pound. Secondarily, the cupric chloride reacts with the cuprous oxide (hydrolytic product) forming cuprous chloride again and basic cupric chloride.Very dilute hydrochloric acid in an atmosphere of carbon dioxide changes the colour of cuprous chloride through green, almost black, to a dark copper brown. The residue in this case consists of metallic copper formed by the decomposition Cu2CI2 = Cu + CuCl? ; but as copper is deposited on the cuprous chloride it protects thls from total decomposition. Cuprous chloride in a solution of cupric chloride, when protected from the action of the air, remains perfectly white even in sunlight, showing that cupric chloride prevents the direct decomposition of the cuprous compound. Perfectly dry cuprous chloride remains unchanged in the air and light has no effect upon it.The method recommended for the preparation of cuprous chloride is to dissolve 42 grams of cupric chloride in 200 C.C. of hydro- chloric acid of sp. gr. 1.175 and 100 C.C. of water, and heat the solution with copper foil on the water-bath until it is decolorised. The solution is then poured into 2 litres of water, the precipitate filtered (in diffused light) and washed, first with dilute sulphuric acid (1 : 20), then with absolute alcohol, drained as dry as possible on the pump and dried quickly in the water-oven. J. McC. Mercury Oxychlorides. By N. TARUGI (Gazxetta, 1901, 31, ii, 3 13-32O).-The author has examined the various oxychlorides of mercury described by different authors, and finds that, without excep- tion, they consist of mixtures, in indefinite proportions, of mercuric oxide and chloride.The three compounds described below are there- fore the first oxychlorides of mercury obtained. The oxychtoride, HgCI,,SHgO, is obtained by adding small cubes of perfectly white statuary marble of a sugar-like structure into satur- ated aqueous mercuric chloride at 15'. After remaining for 15-20 days in diffused light, the liquid deposits the oxychloride in small, yellow crystals which continue to increase in number and size, and are separated, washed with water, and dried in a vacuum. When heated or when boiled with water or alkali solution, the oxychloride decom- poses into i t s constituents, whilst dilute nitric acid converts i t into a white, amorphous powder. If the saturated mercuric chloride solution is diluted with twice its volume of water and treated as before, an oxycidoride of the formula HgC1,,2HgO, is obtained in very thin, black crystals; whilst if three volumes of water are added to the saturated mercuric chloride solu- tion, a compound of the composition HgCl,,HgO, is produced in very thin, red crystals.These two oxychlorides have the same chemical properties as the first described. T, H. P. Alkali Double Nitrite8 of Mercury and Zinc. By ARTHUR ROSENHEIM and KURT OPPENHEIM (Zeit. anoyg. Chem., 1901, 28, 171-1 '74).-Mercuric nitrate, when treated with a concentrated solution of potassium nitrite, dissolves and the solution becomes warm, Mercuric oxide separates out and from the filtrate yellow crystalsINORUANIC CHEMISTRY.21 of potclssiztm ~ c u q nitrib of the composition K3Hg(N02)S,H,0 are deposited. The salt crystallises in rhombic forms [ a : b : c = 0.8594 : 1 : 0*7681] and is soluble in cold water without decomposi- tion. If the solution of this salt contains a slight excess of potassium nitrite, ill-defined crystals of the compound KHg(NO,), are formed. Sodium mercury nitrite, Na,Rg(NO,),, has been prepared by re- crystallising the product obtained by the action of a concentrated solution of sodium nitrite on mercuric nitrate. Potmsium zinc nitrite, K3Zn(N02)5,3H20 is produced when nitrous acid is passed into a solution of potassium nitrite containing zinc hydroxide in suspension. - It forms very hygroscopic, yellow crystals. J. McC. Copper-Aluminium Alloys. By L ~ O N GUILLET (Cmpt.vend. , 1901, 133, 684--686).-The author has prepared various copper aluminium alloys by heating pure cupric oxide with granular aluminium in proportions varying from those which should yield pure copper to those which should yield the alloy CuAl,. By treating the products with acids, he has isolated three distinctly crystalline, definite alloys, Cu,Al, CuAl,, and CuA1, the last being mixed with a small quantity of copper aluminium silicide. Compounds of Aluminium Bromide with Bromine and Carbon Disulphide. 11. By WLADIMIR A. PLOTNIKOFF (J. Buss. Php. Chem. SOC., 1901, 33, 429-432. Compare Abstr., 1901, ii, 31 6).-The compound of the composition AlBr,,Br,,CS,, previously described (Zoc. cit.) by the author as obtained by the action of bromine on a carbon disulphide solution of aluminium bromide, is only formed when the bromine is employed in excess.If, however, to a well- cooled solution of aluminium bromide (1 mol.) in carbon disulphide a quantity of bromine not greater than 1 atomic proportion is added slowly in drops, an almost theoretical yield of an oily compound is obtained, having the composition 2A1Br3,Br4,CS2. When left in a warm place or when shaken repeatedly with carbon disulphide, the oil deposits a brownish, crystalline mass, which, when dry, melts at about 80° and begins to dissociate into aluminium bromide and CS2Br, at about looo; dissociation also occurs under the influence of even traces of moisture. C. H. B. The compound is soluble in ether or benzene. T. H. P. By SAMUEL A. TUCKER and HEBBEBT R.MOODY (J. Xoc. Chm. Ind., 1901,20, 970-971).-1n the electric furnace, aluminium oxide is not reduced by carbon, but if lime is added calcium carbide is formed and this reduces the alumina. Calcium carbide reduces alumina and the yield of aluminium is increased by the presence of free carbon. The heat should not be applied longer than 15 minutes, for after that time aluminium carbide is formed. Periodic System and the Propertiels of Inorganic Compounds. 111. The Solubility of Alums aa a Function of Two Variables. By JAMES LOCKE (Amr. Chem, 2, 1901, 26, 332--345).-The author The Reduction of Alumina by Calcium Carbide. J. McC.22 ADSTRACTS OF CNEMICAL PAPERS, has previously determined the solubility at 25’ of the alums of aluminium, vanadium, chromium and iron severally with ammonium, thallium, rubidium and csesium (Abstr., 1901, ii, 656).When the solubilities of these 16 compounds expressed in gram-mols. per litre of water are plotted as a function of the atomic weights of the tervalent metals, a figure of remarkable regularity is obtained, from a aonsider- ation of which it is evident that the lines joining the solubility points of the successive univalent metals with two given tervalent metals have approximately a common point of intersection. It must be assumed, therefore, that the points representing the solubilities stand in fixed mathematical relation t o one another, Hence it is shown that if the difference in the solubility of the alums of a given tervalent metal with two alkali metals is termed the increment of solubility for the latter,’’ the general law is obtained that the ratio between the in- crements of solubility of the corresponding alums of two tervalent metals for any two alkali metals is constant.’’ The accuracy of this law is fully confirmed by observation.A general equation for the solubility of any of the sixteen alums is deduced from this law, all the terms of which can be referred to two variables, one-applying to the tervalent element in the compound and the other to the univalent metal. Determinations made at other temperatures than 25” indicate that it will be possible to derive a general solubility formula for all temperatures. E. G. The Reaction of Sodium Thiosulphate with Potassium Per- manganate. By A. ALANDER (Zeit. unaE. Chern., 1901,40,574-577).-Both qualitative and quantitative proofs are given that in alkaline solutions the reaction 8KMn0, + 3Na2S20, = 3Na2S0, + 3K,SO, + 8Mn0,+K20 is the principal one, but that a small quantity of the thiosulphate (about 1.3 per cent.) is oxidised only to tetrathionabe. M. J. S. Separation of Iron. By PAUL NICOLARDOT (Compt. rend., 190lj, 133, 686-688).-When ferric chloride is heated a t 125’ it is converted. into a complex compound in which the ratio of iron to chlorine is 1 : l,, whilst the corresponding sulphate is insoluble, Thejron alloy (1 gram), or compound is dissoved in aqua regia, the nitric acid expelled, andl the liquid evaporated to dryness and heated at 125’ for 4 hours. It is! then diluted with water to 500 c.c., heated to boiling, and about 1 graml of ammonium sulphate added.After boiling for about 15 minutes, the very finely divided precipitate is filtered off, If mercury 01’ cadmium is present, the substance cannot be heated a t 125O without; loss, and therefore the liquid is exactly neutralised with ammonia!, mixed with ammonium sulphate, boiled, and filtered. It is again mixed1 with ammonia until a slight precipitate is formed and again boiled, when the whole of the iron is precipitated. Selenates, phosphates, arsenates,vanadates, and molybdates precipitate iron in a similar manner, and the iron is readily separated from the precipitate by fusion withi an oxidising mixture or an alkali. Ferric Oxide and Hydroxides. By OTTO RUFF (Bey., 1901, 34, 3417-3430. Compare Tommssi, Bey., 1579, 12, 1929, 2334).--- The red, colloidal ferric hydroxide may be converted into true hydrates3 C.H. B.INORGANIC CHEMISTRY. 23 by the aid of considerable pressure under water. At a temperature of 42.5", it yields brown ironstone, at 42.5-62*5', gothite, and at higher temperatures, hydrohaematite. Yellow ferric hydroxide is not a true colloid, as its percentage of water under high pressure does not vary between temperatures of 40' and 70". The red hydroxide appears to lose water a t the ordinary atmospheric temperature and pressure and, at the same time, but somewhat more slowly, takes up water and becomes converted into brown ironstone, the only stable hydrate under ordinary conditions. J. J. 5. Influence of the Separation of Sulphur on the Preoipita- tion of Iron Salts.By A. COPPADORO (Gaxxetta, 1901, 31, ii, 21 7-221).-When hydrogen sulphide is passed through an acidified solution of a ferric salt, a precipitate of sulphur is formed containing small quantities of iron compounds, which the most exhaustive washing is incapable of removing. By dissolving the sulphur from the dried precipitate by means of carbon disulphide, however, the author has succeeded in determining the amount of iron enclosed in the precipi- tate. He finds that the quantity of iron is proportional to that of the ferric salt taken and to that of the precipitated sulphur, but is independent of the amount of acid added to the solution and of the time during which the hydrogen sulphide is kept passing through the liquid. If a precipitate of sulphur is produced in a solution of a ferrous salt, for example, by the addition of thiosulphate and acid, the precipitate is found to contain iron.The author suggests that possibly the presence of iron in these precipitates is connected with Graham's observation that when solutions of two colloids are mixed they are precipitated together. Crystallographic Examination of some Luteocobaltic Salts. By TIMOTHBE KLOBB (Chem. Centr., 1901, ii, 970; from BUZZ. Xoc. fianp, Min., 24, 307-322. Compare Abstr., 1901, ii, 103).-Luteo- cobaltic selenate, Co(NH,)6(8e04),,5H?10, prepared by neutralising luteocobaltic hydroxide with selenic acid and slowly evaporating the solution, separates in thick, brownish-yellow, monoclinic crystals [a : b : c = 1.1350 : 1 : 1*4023. ac= 90'35'1. Luteocobaltic sulphate, Co(NH,),(S04),,5H,0, forms lustrous, monoclinic crystals [a : 6 : c = 1.1230 : 1 : 1.4143.Luteocobaltic hydrogen sulphate, 2Co(NH,),(S0,),,5H,SO4, 10H,O, obtained by adding sulphuric acid to an aqueoub solution of the normal salt, crystallises in small, rhombic octahedra [a; : b : c = 0.99913 : 1 : 1.00061 and is decomposed by water. Lu t eoco bal t ic hydrogen selenate, Co (NH,),( SeO,),, H,SeO,, 5 H,O, pre- pared by adding excess of selenic acid to luteocobaltic hydroxide or to the normal selenate, crystallises in triclinic crystals, of ten twinned [a : b : c = 0.84550 : 1 : 0.47285, bc = 88"50', ac = 80°50', ab = 86'47'1, and is not decomposed by water. Luteocobaltic chlorosulphate, obtained from cobalt chloride or sulphate or by treating luteocobaltic chloride withsulphuric acid or a sulphate,is rhombic [a: 6 : c = 0.99855 : 1 : 1.05381.Luteocobaltic chloroselenate is rhombic [a : b : c = 0.99869 : 1 : 1*0563]. Luteocobaltic ammonium sulphate, [Go( N Ha)Jz( SO4),,( NH,),S04, 8H2.0, prepared by crystallising luteocobaltic sulphate in presence of ammonia, T. H. P. ac = 90'18'1.24 ABSTRACTS OF CHEMICAL PAPERS. separates in rhombic octahedra or thick plates. When moistened with water, the crystals become opaque, but in concentrated ammonium sul- phate solution they remain transparent and ultimately dissolve. Luteo- cobaltic ammonium selenate, [ Co( NH3)6]2( Se04),, ( NH4),Se04,8 H,O, obtained by neutralising a solution of luteoco baltic hydrogen selenate with ammonia, is isomorphous with the preceding salt [ a : b : c = 0,95953 : 1 : 1.20241.After the separation of this hydrate, a hydrate crystallising with 4H,O crystallises out in large, monoclinic prisms [a : b : c == 1.4285 : I : 0.64688. Luteocobaltic chloro- ammonium sulphate, [CO(NH~)~]~(SO~),C~,,~( N.H,),SO,,GH,O, prepared by evaporating a solution of luteocobaltic chlorosulphate with a n excess of ammonium sulphate, crystallises in octahedra and is de- composed by water. ac = 94'42'1. E. IT, w. Crystallographic Study of Alvisi's Luteocobaltiammine Per- chlorates. By FEDERICO MILLOSEVICH (Gaxxetta, 1901, 31, ii, 285). -Luteoco bal tiamm onium diperchlorat e chloride, CO( NH3)(jC1(C1O4)2 , forms rhombohedra), orange-yellow crystals, a = 70'41'. Luteocobaltiammonium perchlorate, CO(NH,)~(C~O,),, gives orange- yellow crystals of the cubic system.T. H. P. Researches on Perchlorates. Luteocobaltiammine Per- chlorates and Observations on Metallo-Ammoniums. I. By UGO ALVISI (Gaxxetta, 1901, 31, ii, 289-301).-The author has pre- pared various luteocobaltiammine perchlorates and gives the methods used by him for their analysis. Luteocobaltiammonium diperchlomte chloyide, Co(NH3),CI(C10,),, pre- pared by the interaction of cold saturated solutions of ammonium perchlorate (3 mols.) and luteocobnltiammine chloride (1 rnol.), crystal- lises from water in hexagonal, yellow plates having a pearly lustre; when rapidly heated, i t gradually loses ammonia, and at about 188' explodes, water vapour, chlorine, and nitrogen being evolved and cobalt oxide and chlorides left. Luteocobaltiammoniurn perchlorate, Co( NH,)6(C104)B is prepared (1) by heating aqueous cobalt perchlorate with excess oi ammonium per- chlorate and ammonia and adding potassium (or better sodium) per- manganate until the liquid assumes a n intense, golden-yellow colora- t i o n ; or (2) by heating a solution of cobalt perchlorate with lead dioxide, ammonia, and excess of ammonium perchlorate until the filtered liquid becomes intensely orange-yellow in colour.It crystallises from water in orange-yellow octahedra belonging to the cubic system and with hydrochloric acid yields the diperchlorate chloride just described. Theoretical considerations are put forward by the author as to the mode of combination of nitrogen in the cobaltiammonium compounds with the other elements present.T. H. P. Cause of the Brown Coloration of Ammonium Sulphide in Presence of a Nickel Salt, By UBALDO ANTONY and G. MAGRI (Gaxxetta, 1901, 31, ii, 265--274).-When hydrogen sulphide or ammonium polysulphide solution is added to an ammoniacal solutionIKORGANIC CHEMISTRY. 25 of nickel acetate in quantity insufficient to precipitate all the nickel as sulphide, the precipitate obtained has the composition NiS, and in the latter case is mixed with sulphur. When, however, an excess of ammonium polysulphide is added to an ammoniacal nickel solution, the precipitate formed is of very variable composition but the dark liquid almays contains a sulphide of the composition NiS,. This sulphide, which is obtained as a black powder, is only slightly athacked by hydrochloric acid but reacts vigorously with nitric acid, sulphur being liberated.In an atmosphere of carbon dioxide, it loses sulphur at 300°, being converted into nickel sulphide, whilst when heated in water in presence of air it slowly oxidises, .giving nickel sulphate and sulphuric acid. Hydrogen sulphide solution has no action on it, but it is dissolved by a solution of sulphur or ammonium polysulphides giving a brown liquid. With ammonia solution, it yields an azure-blue liquid containing nickel, but all the sulphur is deposited in a very fine state of division. Measurements of the electrical conductivity of solutions of nickel tetrasnlphide in ammonium sulphide show that the nickel salt is not present in a state of true solution or in a really colloidal condition, the author regarding it as existing in an inter- mediate state.T. H. P. Action of Sodium Thiosulphate on certain Metallic Salts. By FRANZ FAKTOR (Cfiem. Centr., 1901, ii, 878; from Pharrn. Post, 34, 485-487. Compare Abstr., 1900, ii, 598, 627, 688, 691, and 692)- Ammonium molybdate is reduced by a solution of sodium thiosulphate, forming molybdenum trioxideand the hydrate of thedioxide, Mo02,3H,0, whilst sodium tiingst'ate, when warmed with sodium thiosulphate and a small quantity of nitric acid, yields tungsten dioxide, trioxide, and heptoxide. By the action of sodium thiosulphate on uranyl nitrate, a yellow precipitate of uranyl thiosulphate, U0,S20,, is formed ; the thiosulphate, on ignition, yields the green oxide, U,O,. Beryllium thiosulphate, BeS203, 1 1 H20, is prepared from sodium thiosulphate and beryllium sulphate.Quinone is reduced by sodium thiosulphate, forming first quinhydrone and then quinol. By the action of sodium thio- sulphate and hydrogen dioxide on manganese salts, a brown precipitate of manganese hydroxide is obtained ; when treated in the same way, chromates are reduced to chromic hydroxide, cobalt salts give a black and nickel salts a pale green precipitate. E. W. W. By F. MAWROW (Zeit. unorg. Chent., 1901, 28, 162--166).-0n addition of hypophos- phorous acid to a solution of ammonium molybdate in concentrated hydrochloric acid, a bluish-green solution is produced and a violet deposit with a coppery lustre obtained. Thisdeposit is soluble in cold water, giving a green solution, which, on exposure to air, becomes blue. It is decomposed by alkali, forming a green precipitate.It is soluble in con- centrated sulphuric acid with a blue colour, and on dilution a yellowish brown precipitate is formed. Heated on platinum foil, it explodes and leaves a grey residue. Its composition can be represented by On heating an aqueous solution of this, it becomes blue and on evaporating a t 90-95' a blue residue is obtained which is soluble in Two Phosphorus-Molybdenum Compounds. Mo,O,(H,P02)7,3H,O.26 ABSTRACTS OF CHEMICAL PAPERS. water or alcohol with a blue colour and explodes when heated. I t s composition is represented by Mo,O,,(H,PO~)~,H,O. Both these compounds are strong reducing agents, indicating that the phosphorus is present in the condition of hypophosphorous acid, It is doubtful if the formulz given are correct, but it is certain that the substances are not compounds of molybdic acid, but of a lower oxide of molybdenum.The blue solution gives characteristic precipitates with salts of ammonium, barium, lead, and bismuth. J. McC. Behaviour of Hydrochloric Acid Solutions of Metaatannic Acid towards Hydrogen Sulphide, By GUNNER JORGENSEN (Zeit. anorg. Chem., 1901, 28, 140--153).-Hydrogen sulphide gives with hydrochloric acid solutions of metastannic acid, precipitates which vary in composition according to the concentration of the hydrogen sulphide, the concentration of the hydrochloric acid, the temperature, and the time. At first, the precipitate consists for the greater part of metastannic acid mixed with a small quantity of stannic sulphide. When kept, the precipitate absorbs hydrogen sulphide slowly and only after a very long time (two months) does the composition corre- spond with that of stannic sulphide.Increase of the concentration of metastannic acid, or of hydrogen sulphide, or rise of temperature lead to an increase in the proportion of stannic sulphide formed, Increase of the concentration of hydrochloric acid diminishes the absorption of hydrogen sulphide. The rate of formation of the stannic sulphide is extremely slow and it decreases to a far greater extent than the law of mass action would indicate; this is possibly due to the formation 05 s$i&h ' i i ' i A ~ + u ~ . i ~ i ~ + u ~ h~fi. 3.9k9J. Thiocyanates of Quadri valent Titanium. By ARTEUR ROSEN- HEIM and ROBERT COHN (Zeit.anorg. Chern., 1901, 28, 167-170). Compare this vol., ii, 244).-Thiocynnic acid solution dissolves large quantities of titanic acid, and on evaporation of the saturated solution in a vacuum over sulphuric acid a brownish-red, crystallirze powder is obtained which is soluble in cold water, exhibits the reactions of thiocyanates, and possesses all the properties of a titanium salt, It has jh ymqnmition TiO{SCN!,. 2H,O. A solution of titanic acid in thiocyanic acid, when mixed with potass- ium thiocyanate and evaporated over sulphuric acid, gives hygroscopic, deep red, rhombic crystals of K,TiO(SCN),,H20. This is soluble in cold water, but on standing it is decomposed with decolorisation. The corresponding ammonium, sodium, and barium salts have been obtained, but not quite pure.The pyridine compound, (C5NH,),H,TiO(SCN),, as a purple pre- cipitate, is obtained when a concentrated hydrochloric acid solution of pyridine in alcohol is added to a solution of thiocyanic and titanic acids. The precipitate can be recrystallised from the mother liquor and is then obtained in well-defined, bluish-black crystals. The quinoline salt, (C,NHt),H2TiO(SCN),,4H,0, can be prepared in the same way. By recrystallisation from the mother liquor, it separates in deep red crystals, which could not be obtained pure but are probably anhydrous. J. McC.INORGANIC CHEMISTRY. 27 Fluorovanadium Compounds. By PETR G. MELIKOFF and P. KASANEZKY (Zeit. nnoy-g. Chem., 1901, 28, 242-254),-Potassium vanadium dioxyfluoride, VO,F,BKF, when treated with successive portions of hydrogen peroxide, behaves like a compound of the con- stitution VF,(KO),. The fluorine is gradually replaced by oxygen, and orange to red crystalline compounds are obtained containing succes- sively less fluorine, the final product of the reaction being pervan- adate. The corresponding ammonium salt reacts in a similar manner. E, C. R. Preparation of Pure Stibine. By KARL OLSZEWSKI (Rer., 1901, 34, 3592--3593).-With reference to the statements of Stock and Doht (Abstr., 1901, ii, 556), the author points out that the pheno- menon previously observed by him (Abstr., 1886, 977) was a decom- position, not a dissociation, of liquid stibine, and that in his experi- ments air had access to the liquid. Stibine boils at about - 1 8 O , probably somewhat lower than this. A. H. Gold Haloids. By FELIX LENGFELD (Anzer. Chem. J., 1901, 26, 324-332).-Aurous chloride is insoluble in water and dilute nitric acid, and is decomposed by strong nitric acid with production of auric chloride and gold. When nitric acid is added to an ammoniacal solution of the salt, a white, curdy, unstable precipitate is formed. If aurous chloride is added to solution of potassium bromide, gold separ- ates, and a brownish-red liquid is obtained, containing potassium aurichloride and auribromide. It dissolves in a solution of sodium chloride with formation of an unstable double salt, probably NaAuC1,. Aurous bromide is insoluble in water and nitric or sulphuric acid, but dissolves in ammonia with partial decomposition ; on addition of nitric acid to the ammoniacal solution, an unstable precipitate is produced. Potassium cyanide dissolves the salt without decomposition. Potass- ium bromide yields potassium auribromide and gold. With hydro- bromic or hydrochloric acid, it is converted into bromoauric or chloro- auric acid and gold. Both aurous chloride and bromide are slowly decomposed by ether, alcohol, or acetone. Auroso-auric bromide (aurous auribromide), obtained by the action of bromine on gold, is a steel-blue substance, stable in the absence of water, but easily decomposed by water and many organic solvents. Chloroauric and bromoauric acids form the compounds HAuCI,, 3H,O, HAuBr4,3H,0, and HAuBr4,6H,0, but the compounds HAuCI4,4H,O and HAuBr,,5H20 probably do not exist. When solutions of chloroauric acid and silver nitrate are mixed a t Oo, a yellow precipitate is formed which rapidly darkens and decom- poses. If an alcoholic solution of potassium aurichloride is shaken with silver carbonate, silver chloride and auric chloride are produced. E. G.