ii. 46 ABSTRACTS OF CHEMICAL PAPERS. Inorganic Chemistry. Surface Tension and Molecular Complexity of Chlorine. E. MARCHAND (J. Chim. I'hys. 1913 11 573-576).-The surface tension of chlorine has been measured by the method of capillary rise in narrow tubes at temperatures from Oo to 50° and from the results the values of the constant K of the Eotviis formula have been calculated. It is shown that the value of K lies betwesn 2.07 a t Oo and 2.14 at 50° and consequently liquid chlorine is to be regarded as a non-associated liquid. R E N ~ DUBRISAY (Compt. v d . 1913 157 1150-1163).-8 study of the neutralisation of periodic acid by a method already described aad applied to chromic acid (compare A. 1913 ii 388 712). The results which are set out in tables and curves show that periodic acid in solution behavea as a tribasic acid.J. F. S. The Neutraliraation of Periodic Acid W. G. Melting Point of Oxygen. TAD ESTBEICHER (Zeitsch. physikat. Chum. 1913 86 432-434).-The author shows from some resultsINORGANIC CHEMISTRY. ii. 47 previously published (A 1904 ii 477) that in accord with Wahl's suggestion (A. 1913 ii 208) the melting point -227O is most probably the transition temperature of the one modification of solid oxygen into the other. The melting points at different pressures are extrapolated and found to be 0.46 mm. -221*8O 0.87 mm. -219*9O and 1-12 mm. -219.1O. It is shown also that the heat of change of one form of solid oxygen into the other is probably less than that of sulphur which has a value of 2.5H per gram. J. F. S.Revision of the Density of Oxygen Density of the Air at Geneva F. 0. GERMANN (Compt. rend. 1913 157 926-929).- The author has determined the density of oxygen as prepared by heating recrystallised potaesium permanganate and subjecting the gas to liquefaction and subsequent fractional distillation. As a mean of fifteen determinations he finds the weight of a normal litre of oxygen to be 1.42905. I n four of the determinations the oxygen was passed over platinised asbestos a t 400° prior to measuring the density. Two diagrams are given showing the arrangement of the apparatus used. This apparatus has also been used to determine the density of the air a t Geneva in March on two different dates. The mean value found was 1.2930 for a normal litre of air (compare Guye Kovacs and Wourtzel A.1912 ii 636). W. G. The Solubility of Atmospheric Oxygen in Water. TOR CARLSON (Zeitsch. angew. Chem. 1913 26 713-714).-The author has collected the most trustworth7 data obtained by previous investi- gators and also made fresh experiments himself using a modification of Winkler's method with respect to the solubility of atmospheric oxygen irG water at temperatures varying from Oo to 25O. From the results obtained a table is given showing the solubility for every degree between Oo and 25O (a) at 760 mm. ( b ) at (760-0 mm. where f is the vapour pressure of water at the particular tempera- ture. T. 8. P. Formation of Ozone at Various Pressures. H. VON WARTEN- BERG and L. NAIR (Zeitsch. Elektrochem. 1913 19 879-881).-The formation of ozone by a silent discharge has been studied at 8 series of pressures.It is shown that there is a point between 0.5 and 1.0 atmospheres where the formation is a t a maximum. The experiments were continued up to 5 atmospheres pressure and so arranged that the molecules of the gas were always under the influence of the discharge for the same time. Preparation of Sulphur and Sulphates by Heating Sulphites under Pressure. FARBENFABRIKEN VORY. FRIEDR. BAYER & Co. (D.R.-P. 265167).-Sulphur and sulphates can be obtained by heating a mixture of a sulphite and a hydrogen sulphite under pressure when the following reaction occurs 2NaHS0 + Na,SO,= 2N%80 + S + H,O ; for example 150 parts of ammonium hydrogen sulphite 90 p&a J. F. S.ii. 48 ABSTRACTS OF CHEMICAL PAPERS.of ammonium sulphite and 200 parts of water are heated in an autoclave a t 150° for a few hours. J. C. C . The Alloy of Selenium and Iodine. ERNST BECKMANN and ERICH GR~NTHAL (Zeihch. anorg. Chem. 1913 84 97-102. Com- pare Beckmann 2nd Hanslian A. 1913 ii 402; Pellini and Pedrina A. 1908 ii 833).-The tendency to undercooling and t'he difficulty of making cryoscopic observations increase with the con- centration of the selenium. A eutectic point is found a t 50 atomic% and 58O. The eutectic mixture retains iodine even after extraction with solvents until colourless but a t 100-llOo all the iodine can be removed by heat. (2. H. D. The Existence of Compounds of Selenium and Iodine. ERNST BECKYANN and OTTO FAUST (Zeitsch. anorg. Chem. 1913 84 103-112. Compare preceding abstract).-The question whether the dissociation of selenium when dissolved in iodine is to be explained by the formation of loose compounds is not readily answered by thermal analysis on account of the liability to under- cooling.Dilatometric experiments with sulphuric acid ae filling liquid show that the volume change of the eutectic on melting may be calculated from that of the components by the mixture rule. This is also true of the specific volumes whilst the expansion in the formation of the tellurium compound TeI is ten times the experi- mental error. The electrical conductivity shows the eutectic to be a mixture of crystalline iodine and amorphous selenium. Tellurium iodide shows an entirely different behaviour. The molecular weight of selenium Sel0 in methylene iodide is not changed by the presence of iodine 12.The cause of the disso- ciation of selenium in iodine solution must therefore be physical not chemical. C. H. D. Nitrogen Hexoxide and Nitrogen isoTetroxide. F. RASCUIG (Zeitsch. anorg. Chent. 1913 84 115-120. Compare A. 1912 ii 346; Muller ibid. 753).-A reply to Miiller's criticisms as to the point a t which the analysis of the residual solid after the oxygen has evaporated should be made. On washing with liquid nitrogen a dull blue residue is obtained having the composition NO or N,O,. This is nitrogen isotetroxide. The hexoxide is perhaps only stable in presence of an excess of liquid oxygen. C. H. D. The Action of Carbon Dioxide on Boron Sulphide. N. D. COSTEANU (Compt. rend. 1913 157 934-935.Compare A. 1913 ii 694).-Carbon dioxide reacts with boron sulphide in the same way as with silicon sulphide (compare loc. cit.) giving rise to boric anhydride carbon monoxide and sulphur B2S3+ 3@0,=B203+ 3CO + 3s. The reaction commences at 300° and is increased by prolonging the heating and by rise in temperature but is never very rapid being hindered by the formation of a layer of boric anhydride. w. a.INORGANIC CHEMISTRY. ii. 49 The Action of Carbonyl Chloride on the Natural Phosphates and Silicates. J. BARLOT and ED. CHAUVENET (Compt. yend. 1913 157 1153-1155. Compare A. 1911 ii 602).-Carbonyl chloride will attack numerous natural phosphates such as vivianite pyro- morphite uranite and monazite and natural silicates such as thorite gadolinite cerite and zircon yielding in each case the anhydrous metallic chloride.The phosphates are attacked a t tem- peratures between 300° and 500° whilst the silicates require temperatures above 1000° emerald not being decomposed a t 1400O. This reaction forms a ready method of analysis of such minerals and also of preparing anhydrous metallic chlorides from these minerals. W G. Corrosion of Metals by Water. A. T. STUART (J. Ind. Eng. Chem. 1913 5 905-906).-Experirnents carried out with raw and filtered peaty waters showed that although the waters attacked iron t o the same extent in the case of the raw waters a large proportion of the metal removed by corrosion remained in solution possibly in a colloidal state whilst the filtered waters deposited Solubility of M6tale in Water.VICENTE M. ~SNARDI ( A n d . Xoc. Quirn. Argentina 1913 1 214-221).-The course of solution was traced by electrical-conductivity determinations plates of metal 4 cm. long by 2 cm. wide being placed in contact with 70 C.C. of twice distilled water for ten twenty and thirty days a t 18-22O and for one two and three hours at looo. Tin and silver are the least soluble. Copper is more soluble than lead although the latter is more corroded. The solution of lead is proportional t o the time. In the case of zinc the solubility rises very rapidly after twenty days. Tendency of Haloids and Other Salts of the Same Metals to Combine. Fluorides Chlorides and Carbonates. M. AMADOKI ( A t t i R. Accad. Liizcei 1913 [v] 22 ii 366-372. Com- pare A. 1913 ii 216).-The paper deals with the thermal analysis of the systems NaF-Na&O KF-&C03 NaCl-Na&Os and Sodium fluoride and sodium carbonate do not form any com- pound and are not miscible in the solid state.There is an eutectic corresponding with 39 mo1. % of sodium fluoride and 690O. I n the system KF-K,CO the curve of crystallisation of the carbonate descends to an eutectic a t 688O (about 46 mol. % KF) and that of the fluoride to an eutectic a t 682O (about 60 mol. % KF). Mixtures containing 46-60 mol. % KF solidify between these temperatures a compound KF,K,CO being formed which gives with the components two eutectics which melt almost a t the same temperature as itself. Sodium chloride and sodium carbonate do not form compounds and are practically immiscible in the solid state.There is an eutectic at 636O and 59 mol. % NaCI. a ferruginous sediment. w. P. s. G. D. L. KCl-KZCO3. VOL. CVI. ii. 4ii. 50 ABSTRACTS OF CHEMICAL PAPERS. The systam KC1-K2COs is analogous to the preceding; the eutectic temperature is 636O corresponding with 65 mol. % KCl. R. V. S. Crystalline Form of CEsium and Rubidium Nitrates. A. DUFFOUR (Budl. SOC. f r a y . Mi?&. 1913 36 136-143).-Crystals of msium nitrate crystallised from a solution containing also casium dichromate are of two habits on0 as a hexagons1 prism terminated by a hexagonal pyramid and much resembling crystals of quartz; the other as pseudo-cubic forms. The system is rhombohedral with the rhombohedral angle 89O54/ and this pseudo-cubic form is further emphasised by twinning. Rubidium nitrate is orthorhombic with a b c = 0.5789 1 0.7108 the angles here also being very near to those of cubic crystals.The two salta ar0 isomorphous and form mixed crystals. I n both salts the double refraction is feeble and a t a lower temperature it disappears altogether t b crystals then being truly cubic. There is thus a passage from the orthe rhombic through the rhombohedral to the cubic form. Ammonium Peroxide. P. MEL~KOV (Ber. 1913 46 3899).- A correction to the communication of D'Ans and Wedig (A 1913 ii 1051). The compound NH40,H of these two authors is identical with the compound (NR&O2,HzO2 prepared by Melikov and Pissarievski. T. S. P. L. J. S. The Hydrates of Silver Fluoride. A. GUNTZ and A. A. GUNTZ jun. (Compt. rend. 1913 157 977-981).-A reply t o Vanino and Sachs (compare A.1911 ii 884) and a study of the conditions governing the formation of the various hydrates of silver fluoride. Their formation is dependent on the temperature the content of the solution with respect t o hydrofluoric acid and is also a function of the nature of the saturated solution for a metastable or stable phase at the given temperature. The starting material was prepared by saturating concentrated hydrofluoric acid with freshly precipi- tated silver hydroxide concentrating on a water-bath filtering hot and cooling. The crystals obtained were collected and dissolved in distilled water. On evaporating in a vacuum a t loo a neutral solution containing 120 grams of silver fluoride in 100 grams of water and suitably seeding the liquid colourless transparent voluminous crystals of the hydrate AgF,4H20 m.p. 18*5O were obtained having at 1 3 O a heat of solution -4.93 cal. Below 1 8 * 5 O this is the only hydrate stable in neutral solution. A solution containing 170 grams of silver fluoride allowed to crystallise between 18O and 38O gives deliquescent prisms of a hydrate AgF,2H20 m. p. 42O decomposing t o the anhydrous fluoride. Its heat of solution is -1.5 cal. at. loo. The presence of hydrofluoric acid lowers the transition temperature of AgF,4H20 to AgF,2H20 5.5% of acid lowering it t o Oo. On evaporating a neutral saturated solution of silver fluoride at 26-36O hard highly refractive slightly yellow deliquescent crystals are obtained of a hydrate AgF,H,O. The heat of solution of this hydrate is +0*85 cal. a tINORGANIC CHEMISTRY.ii. 51 loo. By repeating this crystallisation a t the ordinaq temperature there is formed round each nucleus of the monohydrate volumin- ous colourless rosettes of a hydrate 3AgF,5H20. Both the latter forms are unstable in the presence of crystals of AgF,2H20 giving this hydrate and anhydrous silver fluoride. Each of these hydrates dried in a vacuum over sulphuric acid yields the anhydrous fluoride in an amorphous form. This can be obtained crystalline in the form of ruby-red cubes by evaporation in a vacuum of a solution of silver fluoride containing 5% of hydro- fluoric acid. It has heat of solution 4.3 ca.1. at 16O. Supersaturated solutions of the above hydrates attack silver more or less rapidly according to the temperature giving crystalline silver subfluoride Ag2F.W. G. Preparation of Metallic Chlorides and Sulphates in a Solid Soluble Colloidal Condition. LADISLAUS KARCZAU (D.R.-P. 263286).-Metallic chlorides or sulphates in a dry colloidal condi- tion can be prepared by the action of thionyl chloride sulphuryl chloride or chlorosulphonic acid on the metallic salts of organic carboxylic acids ; when calcium salicylate is warmed with thionyl chloride the mixture divides into two separable layers the under one consisting of colloidal calcium chloride which can be finally isolated by the addition of ether and dried in a vacuum. F. M. G. M. A Formation of Calcium Carbono-phosphate of the Paleo- lithic Age. P. TEILHARD DE CHARDIN (Conzpt. rend. 1913 157 1077-1079).-An account of a deposit of calcium phosphate in the grotto of Castillo (near Santander) exposed by excavations. The phosphate is found covering corroded calcite and appears to be localised ic a region of persistent dampness.Analyses are quoted showing its composition and examination of the deposit shows the similarity both in site and mineralogical character between these phosphates and t-hose of Quercy (compare Lacroix A. 1910 ii 622) and also the possibility of fixing a date for the commencement of the formation of these phosphates. Crystallographic and Thermal Investigation of t h e Ternary System Barium Chloride Potassium Chloride and Sodium Chloride. HANS GENSKY (Juhrb. Min. BeiZ.-Rd. 1913,36,513-558). -The author has made a detailed examination of the quilibrium relationships characteristic of the above ternary system by means of cooling curves f o r liquid mixtures containing the three chlorides in different proportions and by the investigation of the crystallo- graphic structures presented by the solid products.The data obtained for the three pairs of binary systems indicate that barium chloride and potassium chloride form a compound of the formula BaCl2,2KC1. The melting points recorded itre potassium chloride 7 7 5 O ; barium chloride 9 5 5 O ; BaC?12,2KC1 6630. The eutectic temperatures are 660° for BaC12,2KC1 + KCl and 6 5 2 O for BaC12,2KCl+ BaCI,. At 930° the singly refracting modification W. G. 4-2ii. 52 ABSTRACTS OF CHEMICAL PAPERS. of barium chloride which crystallism out at 9 5 5 O is transformed into a doubly refracting form.This is the usual monoclinic modification stable at ordinary temperatures. Barium chloride and sodium chloride yield freezing-point curves which intersect a t a point corresponding with 39 mols. % of barium chloride and a temperature of 654O. Potassium chloride and sodium chloride form mixed crystals and and freezing-point curve exhibits a minimum at about 50 mols. % of each constituent the temperature of the minimum being 660O. On cooling the mixed crystals they give rise to mixtures of potassium and sodium chloride at temperatures between 300° and 400O. The data obtained in the investigation of mixtures containing all three chlorides indicate the existence of a ternary eutectic a t about 540G. At this temperature the co-existing solid phases are barium chloride the compound BaC12,2KC1 and mixed crystals containing potassium and sodium chloride which represent the limiting mixture for the sodium chloride end of the series.The limiting mixed crystals rich in potassium chloride are resorbed a t a temperature higher than the eutectic. H. M. D. Allotropy of Cadmium. ERNST COHEN and W. D. HELDERMAN (Proc. K. Akad. JVetensch. Amsterdam 1913 16 485-489).-When cadmium is heated for a considerable time a t 70-looo in contact with a solution of cadmium sulphate its density diminishes to a small but mbasurable extent. Observations with a dilatometer containing 360 grams of cadmium which had been previously heated for twenty-four hours in contact with a solution of cadmium sulphate have shown that this change in the density is due to an enantiotropic transformation wh'ich occurs a t 64.9O.The observation affords an explanation of the disintegration which was found by Matthiessen and Bose t o occur in cadmium wires heated ,at 80°. H. M. D. An hydrous Sulphates VII. Cadmium Sulphate with Lithium Sodium and Potassium Sulphates. G. CALCAGNI and I). MAROTTA ( A t t i 12. Accad. Lincei 1913 [v] 22 ij 373-379).-The m. p. of cadmium sulphate is 1000° but there is a marked thermal effect a t 820° indicating a transformation at this temperature. Cadmium sulphate and lithium sulpliate beh'ave like the other sulphahes of bivalent metals hitherto described ; there is an eutectic a t 5 5 1 O (45% C'dSO,). The reactions which occur between the solid components of the system CdS0,-Na,SO are very complicated and numerous and the diagram of the system differs entirely from that given for it by Le Chatelier (Ann.des Mines 1897 [ix] 11 209). Three compounds are formed namely CdSO,,SNa,SO (at 351°) CdSO,,Na,SO a t (551°) and 3CdS04,Na2S0 (at 7 4 6 O ) . The com- pound CdS04,Na2S04 shows a transformation a t about 496O and the compound 3CdS04,Na,S0 has a transformation point a t about 456O. R. V. S.INORGANIC CHEMISTRY. ii. 53 The Behaviour of Lead Carbonate Basic Lead Carbonate and Lead Sulphate in Aqueous Solutions of Alkali Carboa- ates. FRIEDRICH AUERBACH and HANS PICK ( A d . Kuis. Gesundh. Amt. 1913 45 113-165. CompRre Pleissner A. 1908 ii 40; Herz A. 1911 ii 972).-In connexion with the fate of lead paints in the human body the behaviour of lead carbonate basic lead carbonate (white lead) and lead sulphate towards solutions of the alkali carbonates has been investigated. Lead carbonate is transformed into basic lead carbonate by solutions of potassium carbonate whilst the basic carbonate is changed to the normal carbonate by the action of potassium hydrogen carbonate.These two reactions lead to the equilibrium expressed by the equation 3PbC0 + K2C0 + 2H20 L= 2PbC03,Pb(OH)2 + 2KHC0,. The ratio of the potassium carbonate to the potassium hydrogen carbonate concentration is displaced in the direction of higher carbonate concentration as the total alkali concentration increases and in the direction of higher hydrogen carbonate concentration with increasing temperature. The two lead carbonates behave similarly towards solutions of sodium carbonate and sodium hydrogen carbonate as long as the total sodium concentration does not exceed 0.077.N. Above this concentration a double salt basic sodium lead carbonate NaPb2(C03)2-OH is formed the stability of which increases with increasing sodium concentration. This salt is quite different in appearance from the normal and basic lead carbonates being greyish-yellow in colour.The conditions necessary f o r its formation and decomposition were thoroughly investigated. By experiments carried out a t 18O and 37O using solutions in which the total sodium concentration varied up to 0*25W the existence domains of the two lead carbonates and of the double salt were ascertained and the constants calculated for the equilibria expressed hy the equations 2PbCO,,Pb(OH) + 2HC03’ t 3PbC0 + CO,” + 2H20 ; 3[NaPb2(C03)2-OH] + R 2 0 z 2[2PbC03,Pb(OH),] + HC03’+ CO,/’ -J- 3Na’.The constants had respectively the approximate values 32.5 12 and 0.05 a t 18O the concentrations of the substances on the right- hand side of the above equation being in the numerator. From these results the dissociation pressure of carbon dioxide in the change from normal to basic lead :arbonate was calculated t o be about 0.1 mm. Hg a t 18O. Also from these results the authors were able to devise methods for the preparation of pure basic lead carbonate and pure hasic sodium lead carbonate. Solutions of sodium carbonate and hydrogen carbonate which also contain varying quantities of sodium sulphate behave qualitatively towards normal and basic lead carbonate in the same way as sulphate-free solutions ; quantitatively the equilibrium con- centrations are altered in a regular manner.The formation of lead carbonate and a solution of sodium ZPbCO + Na’ + CO,” + H,O NaPb,(CO,),*OH +- HCO,! ;ii. 54 ABSTRACTS OF CHEMICAL PAPERS. sulphate by the action of a solution of sodium carbonate on lead sulphate is not a direct reaction. Basic lead carbonate is first formed and then basic sodium leaci carbonate which in the presence of sufficient lead sulphate is transformed into normal lead carbonate. Lead sulphate and sodium hydrogen carbonate interact according to the equation PbSO + 2NaHC03c PbCO + Na2S04 + CO + H,O from which it follows that lead carbonate suspended in a solution of sodium sulphate can be transformed into lead sulphate by the action of carbon dioxide under pressure.From the various equilibrium constants ascertained in the above experiments the following solubility products (at 18O) were calcu- lated [Pb"j[CO,//] = 10-13; [Pb"]3[C03~~]2[0H1]2= 3*5 x [Pb"]2[Na'][C03'~]2[OH~]=10-~1. T. S. P. The Behaviour of Lead Chromate and Basic Lead Chromate in Aqueous Solutions of Alkali Carbonates. FRIEDRICH AUERBACX and HANS PICK (Arb. Kais. Gemmdh.-Amt. 1913 45 166-190. Compare Lehmann Arch. Hyg. 1893 16 315; Golblum and Stoffella A. 1910 ii 698).-The authors have investi- gated the behaviour of lead chromate basic lead chromate and lead carbonate towards dilute solutions of sodium carbonate sodium hydrogen carbonate sodium chromate and mixtures of the same.I n dilute solutions of sodium carbonate the following equilibrium is formed 2PbCr0 + 2Na2C0 + H,O With increasing dilution and rise in temperature the equilibrium is displaced towards the right-hand side of the above equation. When the equilibrium is approached from the side of the basic lead chromate exactly the same results are not obtained as from the side of the normal lead chromate but this point was not further investigated. PbO,PbCrO + 2NaHC0 + Na,CrO,. The reversible reaction expressed by the equation PbCrO + Na2C0 Z? iPbCO + Nat2Cr04 only takes place in the presence of considerable quantities of hydrogen carbonate in solution otherwise basic lead chromate is formed. The existence domains of the two lead chromates and of lead carbonate in solutions in which the total sodium concentration was 0.05N and O*lN were ascertained a t 18O; the equilibrium constants for the following reactions 2PbCr0 + 2C0," + H20 PbO,PbCrO + CrO," + ZHCO,' ; SPbCO + CrO," + H,O PbO,PbCrO + 2HC0,' ; PbCrO + CO," 2 PbCO + CrO," were found to be respectively 0.057 1-95 and 0.23 the concen- trations of the substances on the right-hand side of the equations being in the numerator.Lead chromate enters into reversible reaction with a solution of sodium hydrogen carbonate according to the equation P bCrO + 2NaHC0 .- PbCO + Na,CrO + CO + H,O ;INORGANIC CHEMISTRY. ii. 55 equilibrium .is attained a t c much lower pressure of carbon dioxide than is the case with lead sulphate (compare preceding abstract) The following solubility products a t 1 8 O were calculated from the various equilibrium constants [Pb”] [CrO,”] = 2 x 10-14; [Pb”]2[Cr04”] [OH’]z= 6 x 10-35 (compare Beck and Stegmuller A 1910 ii 1067).T. S. P. The Amount of Lead given up by Sparingly Soluble Lead Salts to Solutions containing Sodium Hydrogen Carbonate. FRIEDRICH AUERBACH and HANS PICK (Arb. Rais. Gtsundh-Amt. 1913 45 191-196).-Dilute solutions 0-02N and 0*1N of sodium hydrogen carbonate which may also contain sodium chloride and free carbon dioxide dissolve only 0.3-0.4 mg. of lead per litre when shaken up a t 37O with lead carbonate lead chromate or basic lead chromate. It follows that injurious quantities of lead cannot be dissolved from the above compounds by the action of the pancreatic and intestinal juices in the human body. T.S. P. Alkaline Cuprothiosulphates. PIERRE DUTOIT (J. Chirn. Phgs 1913 11 650-673).-The author has studied the titration of copper salts (nitrate and sulphate) by thiosulphates (sodium potassium and calcium) making use of electrical conductivity electric potential and lowering of the freezing point of the solutions to determine the end points. The titrations are carried out in both ways that is adding the copper salt to the thiosulphate solution and by adding the thiosulphate to the copper salt. It is shown that by all methods the titration is complete when 0.9 mol. of a copper salt has been added to 2 mols. of the thio- sulphate or when 2.2 mols. of thiosulphate has been added t o 1 mol. of the copper salt.The author isolates from these reactions the salt 4CuS,0,,3Na,S,03,9H,0 which is the least soluble product formed. In the titrations two inflexion points are noted in the precipitation curves; the first is held to correspond with the formation of the complex Cu,S2O,,9M,S,O and the second to the complex Cu,S20,,5M,S,0,. I f the titration is carried out very slowly it is possible to get producks of the formula Cu,S203,M2S,0,. In these salts the copper is shown t o exist in the anion complex. J. F. S. The Chemical and Physical Nature of Colloidal Hydrous Aluminate Silicates. R. GANS (Centr. Him. 1913 699-712 72&-741).-Aluminate silicates with the composition 3-4 mol. SiO 1 mol. A1,0 1 mol. base (CaO MgO K20 Na,O) are pre- pared by the action of a solmion of alkali aluminate on silicic acid but not by the interaction of alkali silicate and aluminium hydroxide.They are colloidal and very readily dissociated and resemble the zeolitic silicate of soils and the artificial product called permutite. Detailed arguments are given in favour of regarding them as definite chemical compounds rather than as adsorption compounds. L. J. S.ii. 56 ABSTRACTS OF CHEMICAL PAPERS Hypereutectic Alloys of Iron and Carbon. H. HANEMANN (Zeitsch. anorg. Chem. 1913 84 1-23).-The investigation of Wittorf (A. 1912 ii 259) leads to the assumption of several iron carbides Fe,C Fe,C FeC and possibly FeC,. The experiments are open to the objection that the optical examination of the surface in contact with carbon vapour did not give the correct temperature of the molten mass.In the present experiments electrolytic iron and sugar charcoal are fused together stirred with a carbon rod and cast in iron chills after observing the temperature by an optical pyrorueter. The cementite contents obtained by analysing the chilled specimens give points which lie between the curves found by Wittorf and by Ruff (A. 1911 ii 897). The higher the temperature the less accurate is the determination of these points but the assumption of distinct breaks in the curve does not seem to be justified. On quenching from temperatures of 1400O and upwards graphite is always present as a primary product and not as a product of the decomposition of cementite. Microscopical examination shows that the cementite in quenched specimens crystallises radially from the centre indicating that it is formed during cooling.The mechanical separation of- crystais f?om the mother liquor at' a very high temperature as in Wittorf's experiments cannot be used to determine the composition of the solid phase as the separation is always imperfect. Mixtures of molten iron and carbon become very viscous above 1700° and heating to 2500O does not reduce the viscosity. A t 2500O a quenched specimen contains nearly 14% of carbon mostly as graphite entangled in the highly viscous liquid. During cooling the liquid becomes less viscous a t 1700O. Ruff's conclusion that the solubility of carbon again diminishes a t very high temperatures is not confirmed. The dendrites described by Wittorf as a carbide Fe,C are shown to be austenite ils they are converted into martensite by cooling in liquid air and into pearlits by annealing.The supposed carbides FeC and FeC are both regarded as graphite. After treatment with nitric acid the residual carbon is convertible into graphitic oxide and does not contain amorphous carbon as it would do if derived from a carbide. Influence of ailicon on the Solubility of Carbon in Iron. GEORQES CHARPY and ANDRI~ CORNU (Compt. rend. 1913 157 901-903. Compare A. 1913 ii 602).-The authors have deter- mined the solubility of carbon in samples of f erro-silicon containing varying amounts of silicon a t temperatures varying from 600° to 1000°. The results show that silicon causes a gradual diminution in the solubility of carbon in iron the solubility becoming prac- tically nil a t 900° if the silicon content is above 4% and a t 10000 if the silicon content exceeds 7%.At lower temperatures 2% of silicon suffices. W. G. Some Hydrothermal Syntheses. PAUL NIGGLI (Zeitsch. ano9.g. Chem. 1913 84 31-55. Compare Morey and Niugli A. 1913 ii 861).-A principal difficulty in hydrothermal syntheses is the C. H. D.INORGANIC CHEMISTRY. ii. 57 low velocity of reaction when amorphous (labile) substances are present. Such syntheses aim a t the formation of those minerals which can exist a t high temperatures in contact with aqueous solutions. Pressure is only used to maintain the system in the required condition. The products generally result from partial reactions in the amorphous mass and so may contain metastable crystals.The present preliminary syztheses do not therefore necessarily indicate the true stable phases. The apparatus used is that of Baur (A. 1911 ii 991). The materials used are amorphous silica potassium aluminate and silicate and aluminium and ferric hydroxides. The minerals identified are hzmatite potassium zegirite orthoclase and potassium nepheline hydrate. Potassium-aegirite I<,0,Fe203,4Si02 or KFeSi,O is not found in nature whilst its sodium analogue has not yet been obtained synthetically. Another product not definitely identified may be an anhydrous potassium nepheline. Amorphous material is always present. A t 450° hsmatite is obtained together with aluminium felspar free from iron. The presence of hzmatite in natural felspars is not to be attributed to alteration but to separation during cooling from an originally isomdrphous mixture in which ferric iron replaces alumina.The conversion of minerals in eruptive rocks into analcite zeolites and muscovite corresponds with the hydrothermal changes during cooling. Hzmatite occurs in good six-sided leaflets. C. H. D. Alterability and Instability of Cobaltarnmines. NILRATAN DHAR (Zeitsch. anoyg. Chem. 1913 84 224-226).-Great care must bs exercised in the preparation of the cobaltammines. The use of rather more dilute ammonia in the preparation of dinitro- tetramminecobalt chloride than is prescribed in text-books leads to the formation of trinitrotriamminecobalt chloride whilst an increase in the proportions of ammonia and ammonium chloride leads to the formation of nitropenta-amminecobalt chloride.Carb ona totetra-ammineco balt nitrate [Co ( NH3),C0,]N0,,~H20 is slowly decomposed by boiling with water. C. H. D. Ternary Alloys of Nickel-Gold-Silver. P. DE CESARIS (Ganxettn 1913 43 ii 609-620).-These alloys have been investigated by methods similar to those used for nickel-copper-silver alloys (A. 1913 ii 1061) similar results being obtained. The mutual solubilities of nickel and gold a t the eutectic tem- perature are somewhat higher than the values given by Levin (A. 1905 ii 532) gold dissolving about 8% of nickel and the latter about 20% of gold. The space-diagram is described and reproductions of micro- photographs given. T. H. P. The Neutralisation of Chromic Acid. L. MARGAILLAN (Compt. rend. 1913 157 994-995).-h study of the neutralisation of chromic acid by sodium hydroxide the acidity of the solutionii.58 ABSTRACTS OF CHEMICAL PAPERS. being measured by determining the E.M.F. of a ile in which intervenes the half element h j drogenised platinum psolution the other half being mercury 1 calomel I NI10-HCI. The curves given show by sudden alteration in direction the two acid functions of chromic acid; these two points on tlic curves correspond with the end points obtained with helianthine and phenolphthalein as indicators respectively. W. G. 180- and Hetero-poly-salts. IX. Molybdo-phosphites -hypophosphites and -hypophosphates. A RTHUR ROSENHEIM WALTER WEINBERG and JAKOB PINSKER (Zeitsch. anorg. Chem. 1913 84 217-223. Compare A. 1913 i 413).-The oxygen atoms of the alkylarsinic acids are replaceable by MOO or Mo207 residues and the more alkylated the acid the smaller the number of molybdic residues taken k t o the compound.Alkylphosphinic acids are difficult to prepare but phosphorous and hypophosphorous acids may be regarded as derived from them by the replacement of alkyl by hydrogen. An acid solution of ammonium molybdate is precipitated by phosphorous acid and the yellow crystalline precipitate which has the composition (NH,),[HP(Mo~O~)~],~H,O may be used as a delicate test f o r phosphorous acid. The sodium salt N~,[HP(Mo,O~)~],~IH,O is slightly more soluble. It loses all its water a t l l O o . Conductivity measurements show the absence of an acid salt and the salts must be regarded as containing the anion [pr0207)3]." The acid like phosphorous acid is dibasic.The potassium salt is R,[HP(MO,O,)~,~ 1H20 and the yuanidiitium salt Corresponding tungsten compounds have not been obtained the tungsten salts being tribasic as Na3[P(W207)31,1 6H,O. Guaiiidinium molybdoarsenite (CN3H,)3[As(Mo207),],H20 has a similar constitution. A mixturd of sodium hypophosphite (1 mol.) and sodium molybdate (4 mols.) gives precipitates with ammonium potassium or guanidinium chloride. The ammonium salt has the composition NH,[H2P(Mo207)2],6H,0 and these salts may be regarded as derived from the hypophosphite anion PzJ' by replacement of oxygen atoms by molybdic acid residues. Sodium hypophosphate NaHP03,2H,0 (1 mol.) sodium hydroxide (1 mol.) and molybdic acid (6 mols.) yield a yellow crystalline sodium salt Na,[P(Mo20,),],8H,0.PAUL PASCAL (Compt. rend. 1913 157 932-934).-A large number of uranyl salts can be converted into complex salts by the addition of the corresponding alkali salts and according to the acid radicle they possess one of two con- stitutions M,[U02X6] and M,[UO,X,]. The first type where the bivalent radicle UO possesses the maximum index of co-ordination is always very stable resisting hydrolysis and having the reactions of uranium completely masked. (CN3H6)2[HP(M0207)31* C. H. D. Complex Salts of Uranium.IKORGANIC CHEMISTRY. ii. 59 The second type behaves in dilute solution as a double salt and to check dissociation a large excess of the corresponding alkali salt is often necessary. The author has studied the pyrophosphates and the cyanates. On dissolving uranyl pyrophosphate in sodium pyrophosphate solution the freezing point rises to a maximum corresponding with 3Na,P207,(U0,)2P207 and then descends to a minimum a t 2Na4P207,(U0,)2P20T up to which stage the characteristic reactions of uranyl salts are not obtainable.If this solution is evaporated and the gummy residue treated with alcohol a very soluble hygre scopic powder is obtained having the constitution Nad (UO,) 2(P207)31 6H2*. From a solution of sodium pyrophosphate saturated with uranyl pyrophosphate there is precipitated by alcohol a yellow powder Na6U0,[(U02)2(P207)3J,~zH20 which rapidly passes into Na,(UO,),P,O,,H,O insoluble in water. A mixture of alcoholic solutions of potassium and uranyl cyanates gives a microcrystalline precipitate having the constitu- tion K,[UO,( CNO),] very soluble in water gradually dissolving in this solvent to yield an anhydrous orange-yellow double salt KCN0,2U02(CNO),. An excess of alkali cyanate added to the preceding solution gives a golden-yellow precipitate of a double salt.KCNO,UO,(CNO) whilst an excess of uranyl cyanate in slightly alcoholic solution precipitates anhydrous uranyl cyanate as a golden-yellow powder UO,(CNO),. The Isomerism of t h e Stannic Acide. 111. WERNER MECKLENBURG (Zeitsch. nnorg. Chem 1913 84 121-143. Compare A. 1912 ii 355).-The peptonisation of a stannic acid preparation does not depend directly on the size of its primary particles. Readily peptonised prepara-tions are obtained by using nitric acid containing hydrochloric acid. The greater the concentration of the nitric and the less that of the hydrochloric acid the more completely gelatinised is the precipitate.Hydrochloric acid also lessens the oxidation causing the product to contain stannous as well as stannic compounds but in more concentrated solutions the quantity of stannous compound formed is less. If a solution containing stannous salts is heated on the water-bath complete oxidation to the stannk condition takes place violently. The p r e cipitate has sometimes a crystalline appearance but cannot be recognised as crystalline under the microscope. The peptonising action of the reagents examined on the same stannic acid precipitate decreases in the order 22\THC1-+ 22%HC1 -+ 2NHN0 -+ alcohol -+ NH,SO -+ water -j. ether. The preparations have not shown any considerable change in four years.The colloidal solutions obtained by peptonisation have been examined by Tyndall’s method. Hydrochloric acid favours the formation of fine particles. The precipitate contains adsorbed hydrochloric acid. The protective action of this electrolyte closely resembles that of a protective colloid. Alkali hydroxide is also a protective electrolyte. C. H. D. W. G.ii. 60 ABSTRACTS OF CHEMICAL PAPERS. Zirconium Hypophosphite a Zirconium Salt Sensitive to Light. 0. HAUSER and H. HERZFELD (Zeilsch. anorg. Chew. 1913 84 92-94).-Zirconium phosphate Zr(PO,),,H,O is readily obtained by precipitation. Zirconium hypophosphite Zr(OPH20),,H,0 is obtained by add- ing hypophosphorous acid t o a solution of zirconium nitrate until the precipitate has completely redissolved and then adding alcohol.It forms colourless highly refracting crystals which become deep violet very rapidly in direct sudight or in the course of several weeks in diffused daylight without any other perceptible change. C. H. D. Organosols of Metals of t h e Platinum Group. CONRAD AYBERGER (Rolloid-Zeitsch. 1913 13 310-313 Compare A. 19 12 ii 1053 1059).-The preparation of organosols of metallic palladium and platinum is described in which lanolin plays the part of a protective colloid. The lanolin is impregnated with aqueous solutions of salts of bivalent platinum or palladium and the resulting paste triturated with the calculated quantity of alkali metal hydroxide or carbonate. The hydroxides of the platinum metals are then reduced by addition of hydrazine hydrate. On treatment with light petroleum or chloroform the palladium or platinum dissolves together with the lanolin and the colloidal metals can be precipitated from these solutions by the addition of alcohol. Since only a portion of the lanolin separates out on the addition of alcohol this process serves for the concentration of the metal in the preparation. The pasty black substances which are obtained in this way dissolve very readily in chloroform and carbon tetrachloride and are also readily soluble in ethyl ether light petroleum liquid paraffin and fatty oils. The liquid organosols appear to be quite transparent when examined in a very thin layer a. M. u. Organosols of the Hydroxides of Bivalent Platinum and Palladium. CONRAD AMBERCER (Kolloid-Zeitsch. 19 13 13 313-317. Compare preceding abstract) .-Colloidal hydroxides of platinum and palladium in admixture with lanolin as protective colloid have been prepared by the method described in the previous paper. I n thi,s case the addition of hydrazine hydrate is omitted. The pasty substances obtained are readily soluble in light petroleum and other organic liquids which dissolve lanolin. The solution of palladium hydroxide in liquid paraffin has found therapeutic application in the treatment of obesity. The preparation of a palladium oleate organosol is also described. The only modification of the process consists in the substitution of potassium oleate for the hydroxide or carbonate. H. M. D.