INORGANIC CHEMISTRY. Inorganic Chemistry. ii. 35 Nature of' the Decomposition of Hydrogen Peroxide by Light. A. TIAN (Complt. rend 1910 161 1040-1042).-By sub- mttting solutions of hydrogen peroxide in conductivity water t o the action of ultra-violet light from a quartz-mercury lamp the author finds that the reaction for its decomposition is a unimolecular one and proceeds in accordance with the equations H20 = H20 + 0 ; 2 0 = 0,. The decomposition by light therefore resembles that induced by catalysts and is different from the decomposition by heat which is a blniolecular reaction. Under the conditions of the experiment the t 9 . 3 3 2ii. 36 ABSTRACTS OF CHEMICAL PAPERS. reverse action occurred to an extent not exceeding the formation of 0.00005 gram of hydrogen peroxide per litre.w. 0. w. Catalytic Decompoaition of Hydrogen Peroxide in a Homogeneous Medium. ETJGEN SPITALSKY (x Buss. Phys. Chew. Xoc 1310,42 1086-11593C).-A discussion of the nature of catalysis in which it is shown that the velocity and to some extent the direction of all oxidising-reducing actions and particularly that of hydrogen peroxide are greatly affected by the concentration of the hydrogen ions of the reacting mixture and that the condition of the latter at any given moment and whatever the concentration of the hydrogen ions can be completely defined by the effective concentration of the hydrogen peroxide at that moment. I n the catalysis of hydrogen peroxide by dichromnte there is a simultaneous reversible formation of various intermediate compounds of varying stability of which those having the least number of molecules of the hydrogen peroxide attached to the catalyst are comparatively less stable ; hence towards the end of the reaction when the concentration of hydrogen peroxide is less the velocity of the reaction is greatest.The measurements were made gasometrically the experiments being performed in Bredig and Walton’s apparatus (Abstr. 1903 ii 282) and Riesenfeld and Weich’s criticisms (Abstr. 1908 ii 951) of the gasometric method are shown to be entirely without foundation. Many of the results have already been published (Abstr. 1907 ii 338 942). Since chromic acid or the dichromate in acid solution is reduced by hydrogen peroxide to chromic oxide it was necessary in order to complete the experiments on the catalytic decomposition of hydrogen peroxide by chromates to investigate how far this process occurs ia the catalytic experiments described and what is its relation to the activity of the catalyst.I n very dilute solutions all the best ordinary methods for the estimation of chromic acid are wholly inapplicable but the following very accurate and convenient method is employed. To about 10 C.C. of a solution containing 0.0108 mol. CrO 30 C.C. of sodium hydrogen arsenate solution containing 0 004 gram-mols. NaH,AsO per litre are added then 30 C.C. of hydrochloric acid. The mixture is left for twenty minutes after which two drops of methyl-orange or indigo solution are added as indicator and the mixture titrated with an equivalent solution of potassium bromate.The method is recommended as a general one in analysis and as a more accurate and cheaper substitute for iodometric methods. I n the presence of a large excess of hydrogen peroxide chromic acid is at first rapidly reduced although the reaction is not instantaneous after which the hydrogen peroxide is catalytically decomposed under the influence of the chromium tri- and sesqui-oxides but the relative pro- portion of the two last formed finally does not alter with the con- centration or quantity of hydrogen peroxide employed the reaction being one of chemical equilibrium between the two oxides a fact which WRS confirmed by the measurement of the concentration of the hydrogen ions in various mixtures. * The fresh matter i s published also in Zeitsch. anorg.Chem. 1910 69 179-208.INORGANIC CHEMISTRY. ii. 37 Contrary to Riesenfeld’s assumption (Zoc. cit.) it is shown that the form of curves obtained for the decomposition of hydrogen peroxide in the presence of dichromate is best explained by the fact that after the reduction of the dichromate and the establishment of equilibrium between the two oxides the catalytic decomposition of the hydrogen peroxide proceeds without any further change in the catalyst or if any change does occur it is a reversible one and only depends on the con- centration of the hydrogen peroxide at the given moment. The catalytic properties of chromic acid in no way depend on its previous history and the effect of the fresh acid is identical with that of the acid after having been used several times; it is evident therefore that no irreversible changes take place.The curves and the general charac- teristics of the reaction are closely analogous to those of fermenting processes. Z . K. Catalysis of Hydrogen Peroxide. EUGEN SPITALSKY (Ber. 1910 43 3187-3201).-8 reply to the criticisms of Riesenfeld (Abstr. 1908 ii 951). It is also shown that the usual method of determining the velocity of the catalytic decomposition of hydrogen peroxide by measuring the rate at which oxygen is evolved gives the bame results as the direct titration of the hydrogen peroxide with pot a sium permangana t e. T. S. P. The Colours of Colloidal Sulphur. RAPHAEL E. LIESEGANC (Zeitsch. Clwn. l z d . Rolloide 1910 7 307-308).-1f drops of a 20% citric acid solution are placed on a thin layer of a solidified solution containing gelatin and sodium thiosulphate the precipitation of sulphur which results from the diffusion of the acid is accompanied by the development of colour effects.I n transmitted light the colour is a deep lilac-blue The colour is not due to interference but appears to be determined by the size of the colloidal particles. After about twenty-four hours the colour effects disappear and this is supposed to be due to the conversion of S into S,. H. M. D. Reduction of Phosphoryl Chloride by Hydrogen under the Influence of the Electrical Discharge. ADOLPHE BESSON aud L. POURNIER (Compt. rend. 1910 151 876-878. Comp7re Abstr. 1910 ii lZl).-When the vnpour of phosphoryl chloride mixed with pure dry hydrogen is submitted to the action of the silent electrical discharge the armatures slowly become coated with a yellow solid.If this is removed by hot water and dried in a vacuum it is obtained as an easily oxidised reddish-yellow powder having the composition P,O. It appears to be identical with the oxide obtained by the action of phosphine on phosphoryl chloride in presence of hydrogen bromide (Abstr. 1898 ii 216; 1901 ii 502) the existence of which has been questioned by other investigators. The action is stated to take place in accordance with the equation 2POC1 + 4H2 = P20 + 6HCl+ H,O. The reaction was carried out in an apparatus of fused silica- which is described in detail. The use of glass for the purpose is inadmissible w. 0 w,ii. 38 ABSTRACTS OF CHEMICAL PAPERS. Safety Explosives Employed in Mines.J. TAFFAKEL (Compt. rend. 1910 151 873-876).-It is shown experimentally that the gases formed by the detonation of safety explosives of the securite type contain a much greater proportion of carbon monoxide when the charge is enveloped in the usual paraffined wrapper than when the explosive is uncovered or wrapped in ordinary paper or asbestos. The presence of the secondary flame accompanying explosion appears to depend largely on the presence of an oxidisable envelope or of coal dust. It is not seen when asbestos paper is employed and the use of the latter therefore enables heavier charges to be fired with safety. w. 0. w. Bromo- and Hydrobromo-derivatives of Silicon. ADOLPHE BEYSON and L. FOURNIER (Compt. rend. 1910 151 1055-1057).- Four kilograms of the crude product arising from the action of hydiogen bromide on silicon at a red heat were fractionally distilled under diminished pressure.About 90% of the liquid consisted of silicon tetrabromide the remainder containing tribromosilicomethane with a small quantity of dibromosilicomethane SiH2Br2 a spontaneously inflammable liquid b. p. about 75" and probably bromosilicomethane SiH,Br b. p. 30-40". The product of the action of the silent eIectric discharge on tri- bromosilicomethane yielded on fractionation silicon tetrabromide hexabromosilicoethane colourless crystals m. p 9 5 O b. p. 265O an octabromide Si,Br8 crystals m. p. 133" a decabromide Si4Brlo m. p. 185" (decomp.) together with an uninvestigated yellow residue. Silicon tetrabromido differs from the tetrachloride in not under- going reduction when mixed with hydrogen and submitted to the silent electrical discharge.w. 0. w. The History of " Potaah I' and of its Name. EDMUND 0. VON LIPPMANN (Chem. Zeit. 1910 34 1217-1219 1226-1338 1335-1 237).-Historical. L. DE K. The Isomorphism of Potassium and Sodium Compounds. ERNST SOMMERFELDT (Zeitsch. anorq. Chenz. 1910 €39 47-51).-The question of the possible isomorphism of potassium and sodium compounds has been studied in the double sulphates blodite Na,Mg(S0,),,4H20 and leonite K2Mg(S0,),,4H,O the crystals being examined by the ultramicroscope. Crystals of blodite become turbid if containing more than 0.8% K,SO the number of particles increasing with the time whilst leonite only shows similar particles when as much as 11.6% Na,SO is present.According to Osttvald the crystallisation of a supersaturated solution is only brought about by a foreign salt if the latter is isomorphons with the salt in solution. The crystallisation of a supersaturated solution of potassium sulphate on the addition of a crystal of ammonium sulphate may be explained by the dissolution of the crystal displacing potassium sulphate. I n accordance with this explanation a readily sduble salt such as potassium iodide alsoINORGANIC CHEMISTRY. ii. 39 brings about the crystallisation of potassium sulphate. criterion may therefore sometimes give misleading results. Ostwald’s C. H. D. A Characteristic Behaviour of Alkali Phosphate. ERNST SALKOWSKI (Zeitsch. physiol. chem. 1910 69 475-478).-It was noticed that the addition of sodium hydroxide to an American meat juice caused it to set into a crystalline mass.When exposed to the air the juice deposits crystals of potassium dihydrogen phosphate and the addition of sodium hydroxide no longer causes crystallisation to occur in the liquid. W. D. H. Some New Compounds of Nitrogen and Hydrogen with Lithium. I. FRANZ W. DAFERT and R. MIKLAUZ (~ilonc6tsh. 1910 31 981-996. Compare Abstr. 1909 ii 882).-Metallic lithium com- bines with pure dry nitrogen a t the ordinary temperature amorphous lithium nitride Li,N being formed ; the presence of relatively small quantities of oxygen or hydrogen however will completely prevent the reaction. Lithium is not attacked by dry air at the room temperature. Lithium nitride has m.p. 840-845’; the fused substance rapidly attacks iron nickel copper platinum porcelain and other silicates. At 870’ i t will eat through any containing vessel. It penetrates through vesseIs of magnesia as through a filter. When heated at 220-250’ in a current of hydrogen a new com- pound trilithiumanzmoniun Li,NH is formed from the amorphous nitride. If this compound is heated above 340’ in an atmosphere of hydrogen part of its hydrogen is expelled to be absorbed again on cooling. If the temperature is raised to 480’ pure trilithiumarnide Li3NH2 is formed. This latter compound is also formed directly by the action of hydrogen OD crystallised lithium nitride which has been obtained by heating lithium a t 460’ in a current of nitrogen; the reaction is so vigorous that the amide is obtained as a fused mass.Trilitbiumammonium is an extremely hygroscopic substance evolv- ing hydrogen and ammonia under the action of water. Trilithium- amide possesses similar properties ; it is not sensitive towards light. T. s. P. Photo-Halides. 11. WILLEM REINDERS (Chem. mkekbkad 191 0 ’7 993-1005. Compare Abstr. 1910 ii 1062).-The author advances arguments in favoiir of the theory that the photo-halides are absorp- tion compounds of colloidal silver and normal halides. He does not consider that there is any evidence to support the theory that they are sub-halides. A. J. W. ‘( Ripening ’’ of Silver Halide Emulsions. RAPHAEL E. LIESEGANG (Zeeitsch. physikal Chem. 1910 75 374-377).-The increase in size of the particles of silver halides the so-called “ripening” of photographic plates may be accounted for in two ways (2) by the association of small particles a process analogous to the coagulation of colloids and (2) by the gradual increase in sizeii.40 ABSTRACTS OF CTXEMICAL PAPERS. of the larger particles a t the expense of the smaller owing to the greater solubility of the latter. The first process must be associated with a kind of diffusion which could not take place in a solidified gelatin film and the observation that ripening proceeds rapidly in such a dried film supports the latter view Further evidence in favour of this explanation is adduced. G. S. Calcium Silicates in Cement. L. Sz. SZATHMARY (Zeitsch. Kryst. Min. 1910 48 448; from Vegy6sxeti Lapok 1907 2 No. 7 and Bdnydsxuti 5s Kohdszati Lapok 1908 46 635-652).-Several calcium silicates have been prepared synthetically but some of them are probably of the nature of solid solutions.Which of these are present in cements the author leaves undecided. The calcium aluminate 2Ca0,AI,03 is probably a constituent of cement. In- cidentally the following analysis is given of [impure] wollastonite from Csiklova Hungary. SiO,. CaO. Fe,O,,. GO,. H,O. Total. 44.95 49.49 0'58 4'31 0.58 99.91 L. J. S. The Alloys of Tellurium with Cadmium and Tin. MATSUSUKE KOBAYASHI (Zeitsch. ccnorg. Chem. 1910 69 1-9; Mem. Coll. Sci. Xng. Kyoto. 1910 ii 353-363).-Mixtures of cadmium and tellurium fused in glass or porcelain tubes in an atmosphere of carbon dioxide have been submitted to thermal analysis.The freezing-point curve has a maximum at about 1041° corresponding with the compound TeCd but it is not possible to prepare this compound in a pure condition under ordinary pressure owing to the volatility of cadmium. The two eutectic points lie so near to the freezing points of cadmium and tellurium respectively as to be indistinguishable from them. The freezing-point curve of mixtures of tellurium aud tin also has a maximum at 780° corresponding with the formula TeSn. The eutectic point between this cornpound and tellurium lies at 393" and 86% Te. The compound TeSn has D 6.472 and does not form solid solutions with tin or tellurium (compare Fay Abstr. 1907 ii S80; Biltz and Rlecklenburgh Abstr. 1909 ii 1022). The thermal results are confirmed by microscopical examination.C. H. D. Falk's White Lead. JULIUS F. SACHER (Chern. Zeit. 1910 34 1261-1263. Compare Abstr. 1910 ii 712 1067).-When lead carbonate lead oxide and lead acetate are kneaded together the product is not 5PbC03,2Pb(OH)2,Pb0 but a mixture in varying proportions of normal lead carbonate and white lead. The mixtures only become white when air containing carbon dioxide has access otherwise n yellow or red colour persists. True white lead examined microscopically is always amorphous and pure normal lead carbonate is always crystalline different preparations diBering only in the size of their crystals.INORGANIC CHEMISTRY. ii. 41 The decomposition of white Iead by hydrogen snlphide ohsewed by Falk carbon dioxide being liberated is due to the presence of lead acetate the acetic acid set free decomposing a further quantity of carbonate.C. H. D. Formation of Colloidal Copper. ANDR$ RASSEKFOSSE (Bull. Acad. my. Belg. 1910 738-741).-Copper sulphate in sulphuric acid is reduced by various organic liquids forming at first probably violet cuprous sulphate and finally colloidal copper. When a few drops of alcohol are placed in a solution of copper sulphate in sulphuric acid a violet zone which gradually becomes brown is formed between the two liquid layers. On gently warming i t disappears to re-appear on cooling but a t 45' it becomes yellow. The solution is stable and is not affected by filtration or by treatment in a centrifugal machine but when submitted to an electric current deposits a deep reddish-brown granular pulverulent copper quite different from the coherent deposit obtained from a solution of copper sulphate in sulphuric acid.Similar results with some difference in details are shown when methyl alcohol ether acetone acetic acid or chloroform is added to such a solution and a solution of copper phosphate in phosphoric acid shows similar behaviour. T. A. H. Corrosion of Copper and Iron Alloys by Water Con- taining Salt and Air. Oxidation of Copper 4t High Temperatures. WILLEM P. JORISSEN (Zeitsch. angew. C'hsin. 19 10 23 2305-2306).-A specimen of delta metal from the keel of a ship which had been attacked by sea-water was found to contain 41.1% of cuprous oxide no cupric oxide being present; the zinc had completely disappeared. Cu prous oxide was also formed from copper which had been kept for two years either in sea-water or in a solutiori of sodium chloride of the same chlorine content.When copper is partly immersed in distilled water cuprous oxide is formed on the submerged part whereas cupric oxide is formed on the other part. Cuprous oxide is formed when air is continually circulated over copper heated to a temperature above that at which dissociation of cupric oxide takes place. Analyses are given of (a) a piece of iron which had been in the ground for years ( b ) iron used to protect a copper alloy in a pump from corrosion ( c ) iron used for the same purpose in a ship's condenser (d) some Dutch bronze cannons fished up from the Straits of Messina. T. s. P. Occluded Gas in Alloys of Copper. G. GUILLEMIN and B. DELACHANAL (Compt.rend. 1910 151 881-883. Compare Abstr. 1909 ii 144).-Results of analyses are given showing the total volume of occluded gas and the percentage of carbon dioxide carbon monoxide methane hydrogen and nitrogen in different specimens of brass bronze aluminium bronze phosphor-bronze and tin. Forgeable brass contains hydrogen cnrbon dioxide and carbon monoxide ; the latter predominates in unsound ingots containingii. 42 ABS'I'PACTS OF CHEMICAL PAI'ERS. blow-holes. The presence of hydrogen does not impair the mechanical properties of brass The small amount of gas in phosphor-bronze consists chiefly of carbon dioxide and hydrogen. Commercial tin contains hydrogen carbon dioxide and carbon monoxide. w. 0. w. The Tempering of Bronzes. L. GRENET (Compt. Tend. 1910 151 S70-87i).-The connexion between the temperature at which bronze is annealed after tempering and the hardness of the product appears to be the same as in the case of steel.w. 0. w. The Formation of Cuprous Iodide from the Physico- chemical Standpoint. P. P. FEDOTBEFF (Zeitsch. anorg. Chem. 1910 69 22-37).-The reaction occurring between copper and iodine in presenceof water has been studied Cu + I - CuI Z CuI + I. Equilibrium at 20" is only reached after fifteen to thirty hours' shaking. The quantity of iodide dissolved increases with the proportion of iodine. I n solutions in equilibrium with solid iodine the concentration of the dissolved iodine increases with that of the copper. Isothermals for the two systems (cuprous iodide and iodine respectively as solid phase) are drawn and it is shown that in all the saturated solutions the ratio Cu 41 is maintained.The range of composition within which cupric iodide is stable is increased by raising the temperature. Assuming that the complex ion present is I'3 the ratio 1'3/1' at 20" = 0,964. As the cation does not enter into the equilibrium the general conclusion is drawn that in aqueous solution of iodine in iodides the iodine is equally divided between the simple and the complex ions. The formula of cupric iodide in a solution saturated with iodine is Cu<' The number of complex ions increases with rising temperature. The solubility-product Cu' x I' = L = 5-06 x 10-12 (Bodlander and Storbeck Abstr. 1903 ii 607) so that K = ( C U ' * ) ~ ( I ' ) ~ / L ~ ( I ~ ) = The potential of the iodine electrode a t 20° is e = 0.900 + O-O2910g(I,)/(I')2; and of the Cu"/Cu' electrode a t ZOO e = 0.476 + 0.05 810g(Cu")/( CU').I,' 4.18 x 10-14. C. H. D. Double Sulphates formed by Lanthanum and Cerium Sulphates with the Alkali Sulphates. BARRE (Compt. Tend. 1910 151 87 1-873).-The solubility of lanthanum and cerium sulphates in aqueous solutions of potassium sodium and ammonium sulphates at different concentrations has been determined. The existence of the following double salts has been revealed by the solubility curves and confirmed by analyses La2(SO,),,K2SO4,2H2O ; La2(S04),,5K,S04 ; La,( SO,),,Na2SO4,2H20 ; La2(S0,)s,5(NH4)2S0,. C'e,(S04)3,K2S04,2H,0;2Ce2(S0,)3,3K2S04,8H,0; Ce2(S04)3,5K,S04 ; Ce,(S04),,Na2S04,2H20 ; La,(S04)3,(NH,)2S0,,2H20 ; 2La2(S04),,5(NR4)2SU ; Ce~(S04)39(NH4)~S04,sH~0 ; Ce2(S04),,5(NH4)2S04.INORGANIC CHEMISTRY.ii. 43 Lanthanum and cerium sulphates are more soluble in a solution of ammonium sulphate than in one of sodium or potassium sulphnte. w. 0. w. Nature of the Oxides Causing the Colour of Oriental Sapphires. AUGUSTE VERNEUIL (Compt. rmd. 1910 151 1063-1066).-The opinion that the colour of oriental sapphires is due to ferric oxide is based on analyses by Vauquelin Forchhammer and others. The author having previously described the preparation of artificial sapphires from fused alumina coloured by ferric and titanic oxides (Abstr. 1910 ii 212) now shows that the latter is present in the natural stones ; three sapphires of different origin being found to contain 0*03-0.058% of titanic acid.The conclusion is drawn that the colour of sapphires is due to titanium present as an oxide or as a titanate of iron. Catalytic Properties of Asbestos. P. A. TSCHEISHWILI (J. Russ. Phys. Chem. Soc. 1910 42 856-862).-An acid- solution of N'/20- potassium permanganate filtered through asbestos is reduced in each of the first eight filtrations after which it is no longer reduced on filtration. Ten C.C. loses in t h i s way 0.025 C.C. for each filtration and the results obtained are exactly the same for asbestos from various sources and whether it has only been treated with acids and water or whether it has been ignited before such treatment.* The precipitate formed on the asbestos has no catalytic influence on the reduction of the permanganate and it does not consist of manganese peroxide.It is much more convenient to use asbestos in the filtration of perman- ganate solutions and allow 0.025 C.C. as correction than to use glass wool through which a clear filtrate cannot be obtained. V. ANDSTROM (Zeitsch. anorg. Chem. 1910 69 10-21. Compare Moody Trans. 1906 89 720; Friend Proc. 1910 26 179; Idamberti and Thomson Trans 1910 9'7 2426). -Thin strips of soft iron are enclosed with water (the oxygen and carbon dioxide in which have been estimated) in flasks closed by ground stoppers sealed by mercury. After shaking for a definite time the rust formed is collected on a filter that present as an incrustation on the strips being added and the iron in the filtrate is also estimated. The quantity of iron removed by corrosion is independent of the proportion of carbon dioxide in the water and is proportional to the amount of dissolved oxygen the ratio of iron t o oxygen corresponding with Fe,O,.The dissolved portion of tho iron corresponds with Fe(HCO,),. When very little oxygen is present but the solution contains large quantities of carbon dioxide corrosion only proceeds very slowly. The results are interpreted as indicating that rusting is principally due to the action oE oxygen probably with intermediate formation of hydrogen peroxide. C. H. D. Dissociation of Ammoniacal Ferrous Chlorides and the Formation of Ferrous Nitride. FERNAND GIRARDET (Bull. Soc. chim. 1910 [iv] 7 1028-1034. Compare Lang and Rigaut Trans. 1899 '75 883; Jackson and Derby Abstr.1900 ii 59ti).-The w. 0. w. Z. K. The Rusting of Iron. These results are shown in the form of curves.ii. 44 ABSTRACTS OF CHEMICAL PAPERS. interaction of ammonia with ferrous chloride is reversible between -18O and 3509 and a study of the tensions of dissociation of the products indicates that two compounds of this type exists represented by the formuh FeC12,6NH3 and FeCI2,2NH3. The former is stable only at low temperatures. Above 350° the chloride is decomposed and ferrous nitride is formed the optimum temperature for this reaction being 480'. I n preparing ferrous nitride it is best to heat the chloride to 480a and then introduce ammonia gas. F u l l experimental details and tables of dissociation tensions are given in the original. T. A. H. Chemical Constitution of Iron-pyrites and Pyrolusite.L. BENEDEK (Zeitsch. Kryst. Min. 1910 48 447-448; from Magy. Chern. B'olySirat 1908 14 85-81).-Iron-pyrites when heated to a red-heat in an atmosphere of carbon dioxide loses kalf of its sulphur and ferrous sulphide (FeS) remains. At 300-400' in water vapour half the sulphur is also lost after which the ferrous sulphide is decomposed with the production of ferrous oxide and hydrogen sulphide. These reactions indicate in the author's opinion that the iron in iron-pyrites is ferrous. Pyrolusite (Mn02) when heated in an atmosphere of hydrogen passes first into Mn203 at 280' into Mn304 and finally into MnO. Preparation of Tantalic Acid from West Australian Fergusonite. klodium Tantalate. EDGAR WEDEKIND and W. MAASS (Zeitsch.angew. Chern. 19 10 23 2314-231 6).-A specimen oE fergusonite from West Australia contained 5422% Ta,O 1.80% Cb2.0 27.73% Y203 3.20% Ce203 and 0036% Tho and owing to t h e high tantalum content the authors have used it t o prepare tantalum pentoxide and derivatives. The acid and basic constituents were separated essentially according t o the method of Krdss and Nilson (Abstr. 1887 706) the residue then consisting essentially of tantalic columbic and titanic acids; 7.5 grams of the residue were then care- fully added to 15 grams of sodium carbonate fused in a platinum crucible after which a further 4 grams of sodium carbonate and 1.6 grams of sodium nitrate were added. The cold fusion was extracted with 750 C.C. of water filtered and left overnight; 6-4 grams of sodium tantalnte crystallised out.The finest crystals were picked out the remainder dissolved in much hot water and after cooling the tantalum precipitated as tantalic acid by means of carbon dioxide (compare Weiss and Landecker Abstr. 1909 ii 942). The tantalic acid was dried at looo mixed with a little fuming nitric acid evaporated to dryness and then converted into pentoxide by ignition. Sodium tantalate forms colourless hexagonal crystals which give an alkaline solution. Analysis gives Na,Ta,O,,,H 1 2*32H,O agreeing with Rose's empirical formula Na,Ta,0,,25H20 ; 10*5H20 are lost at lO5-1lO0 a further molecule a t 190° and the last molecule in the blow pipe. It is probably an acid sodium salt of-a tritantalic acid of L. J. S. tantalate obtained on heatingbeing a hexatantalate.T. 8. P.INORGANIC CHEMISTRY. ii. 45 Gold 'l?elhtides. GIOVANNI PELLINI and E. QUERCIGH (Atti R. Accad. Lincei 1910 [v] 19 ii 445-449).-The freezing-point curve of the system gold-tellurium has been determined the alloys being fused in an atmosphere of carbon dioxide. The curve has a single maximum corresponding with the compound AuTe melting a t 464'. There are two eutectic points at 12 and 47 atomic % Au and a t 416" and 447" respectively. There is no indication of the formation of solid solutions. The compound RuTe occurs in nature as calaverite. As such a compound is not obtained by the action of tellurium on solutions of gold salts it is probable that the mineral has been formed by a process of fusion Halogen Compounds of Rhodium. Mlle.(3. GOLOURKINE (Bull. Xoc. china. Belg. 1910,21 388-396).-Spongy rhodium readily dissolves in a solution of hydrochloric acid saturated with chlorine or in a solution of hydrobromic acid saturated with bromine giving the corresponding halogen compounds. The preparation of the bromine compounds is described. One gram of spongy rhodium is heated with 24 grams of 40% hydrobromic acid containing 7 grams of bromine in a sealed tube at SO-looo for forty-eight hours. The contents of the tube are distilled under diminished pressure the residue treated with a little water and again distilled to remove all hydrobromic acid. The bromide remain- ing is dissolved in water and the solution evaporated in a vacuum over sulphuric acid the final drying of the product being carried out in a vacuum over fused potassium hydroxide. If the spongy rhodium used contains zinc or bismuth the resulting impure bromide is reduced in a current of hydrogen ; the zinc or bismuth bromide sublimes leaving pure rhodium from which the pure bromide is obtained by the treatment already detailed. Rhodium tribromide RhBr,,%H,O is a blackish-red substance which is very soluble in water. When heated at 100-140° it loses not only water but also hydrogen bromide. With potassium hydroxide the solution gives a precipitate of the hydroxybromide Rh(OH)2Br,2H20 and from the filtrate dark red crystals of potassium rhodobromide K,RhBr5 are obtained. Potassium aodium rubidium ammonium and bccrium rhodobromides are prepared by adding a slight excess of rhodium bromide to the solutions of the bromides of the metals mentioned. The solution is evaporated to dryness and the excess of rhodium bromide extracted with slcohol leaving the rhodobromide undissolved. They possess the general formula M',RhBr5 and are readily soluble in water; most of them are dark red in colour the sodium salt being brick-red. A warm concentrated solution of potassium iodide produces a black precipitate of rhodium iodide RhI from a solution of rhodium bromide. Dilute solutions give no precipitate in the cold. C. H. D. T. S. P,