INORGANIC CHEMISTRY. 81 Inorganic Chemistry. Atomic Weight of Iodine PAUL KOTHNER ancl E. AEUER (Annulon 1904 337 123-169).-A preliminary account of this determination of the atomic weight of iodine has already been published (Abstr. 1904 ii 556). I n the present paper the earlier work on this subject is discussed more especially the recent work of Ladenburg. A detailed account of the methods employed in purifying the material and of the apparatus used in the various experiments is given. Tn discussing the results it is pointed out that the new value for the atomic weight of iodine 126.03 (H = l ) agrees more closely with the values given by Ladenburg (126.008) ancl by Scott (126.01 and 126.03) than with Stas's mean value (1 2590). K. J. P. 0. Revision of the Atomic Weight of Iodine.GREGORY P. BABTEB (J. Amw Chern. h'oc. 1904 26 1577-1595).-Determinations of the atomic weight of iodine have been made by converting a known weight of pure silver into silver iodide. The silver was dissolved in nitric acid and the solution was diluted with water and afterwards boiled to remove nitrous acid and nitrogen oxides. The solution was now treated with an excess of ammonia and afterwards with dilute solution of pure ammonium iodide prepared from highly purified iodine. Great pre- cautions were taken in the purification of the materials and the collec- tion of the silver iodide. Two samples of iodine were employed one of mliich was collected in four fractions the element being liberated from hydriodic acid by the action OF potassium permanganate which was82 ABSTRACTS OF CHEMICAL PAPERS.added in four successive quantities distillation being carried out after each addition. The average result of 15 experiments gave the atomic weight of iodine 126,973 (hg = 107*930) a result which is probably slightly too low. The very small differences in the results show that the samples of iodine were identical and lead to the conclusion that no new halogen of higher atomic weight than iodine existed in the material employed in the investigation. I n another series of experiments the ratio of silver to iodine was investigated by ascertaining the exact weight of silver required to combine with a weighed quantity of iodine. The average of three determinations gave the atomic weight of iodine 136.977. In a third series of experiments four determinations of the ratio of silver iodide to silver chloride were made by converting a weighed quantity of pure silver iodide into the chloride by heating it in a quartz crucible in a current of chlorine.Two determinations were made by heating silver iodide in a current of car5on dioxide and bromine and afterwards in chlorine. From the ratios obtained in these six experiments the average atomic weight of iodine was found to be 126.975 (Ag = 107,930 ; C1= 35.467 ; 0 = 16). The close agreement of this 1 esult with the average of the results of the first two series indicates that the atomic weight of iodine is 126.975 and is additional evidence in favour of the value 35.567 for the atomic weight of chlorine which was found by Richards and Wells in an investigation hitherto unpublished.In connection with the experiments described in this paper the sp. gr. of pure fused silver iodide was determined and found to be 5.674 at 25'14'. E. G. Flowers of Sulphur and Sublimed Sulphur. A. DOMERGUE (Anw. China. anal. 1904 9 445-449).-According to the author the name of YZowem of suZphu~ should be restricted to samples which when newly made contain a minimum of 33 per cent of sulphur insoluble in carbon disulphide. If containing less than this the name sublimed sulphur should be applied. L. DE K. Contact Method for the Manufacture of Sulphuric Acid. FRIEDRICH w. KUSTER [with FRANKE and GEIBEL] (Zed. ccnorg. Chem. 1904 42 453-469).-1n order to maintain a uniform current of the mixture of sulphur dioxide and oxygen the gases were passed from a copper gasholcler which is described in detail.Platinum vanadium pentoxide and ferric oxide induce f o r equal temperatures the same state of equilibrium bet ween sulphur dioxide oxygen and sulphur trioxide. Platinum is the most effective of the three catalysts mentioned and is the only one of technical importance. Water has considerable influence on the catalytic activity of ferric oxide and vanadium pent oxide. The catalytic effect of ferric oxide is iiiipaired by its sensitiveness towards mechanical and chemical influences. The effect of increasing amounts of arsenious oxide on it is gradually to destroy its activity. Vanadium pentoxide is not however so sensitive in this respect. A. McK.I NO K G AN 1C C H E M ISTR Y.83 Raschig’s Theory of the Lead-chamber Procem. EDWARD DIVERS (J. SOC. Chem. Ind. 1904 23 (24) 1178-1 182).-The author gives a &sum5 of the facts adduced in support of the theories of Lunge and of Raschig and suggests a modification of Raschig’s theory. According to this view there is no significant quantity of any oxide of nitrogen in the actually gaseous part of the contents of the chamber in those regions where the change is in active operation but nitrogen peroxide is present where the activity is much leas. On entering t’he chamber from the Glover tower the whole of the oxides of nitrogen together with sulpliur dioxide and water condense to form a mist ; sulphur dioxide oxygen and water then condense together upon the liquid particles of the mist and unite wider the catalytic influence of the nitrous acid present in the particles;.In Raschig’s theory of the chaniber process there is no place for nitrososulphuric acid although according t o Lunge its presence is certain and of prime importance. I n the author’s opinion the nitrous acid in the liquid iuist particles is present in the form of nitroso- sulphuric acid which is to be looked on as a mixed anhydride of pyrosulphuric and nitrous acids and the mechanism of the process call be expressed by writing nitroso-sulphuric acid for nitrous acid in Raschig’s equations 2H,O + 20N.0*S03H + 2S0 = 2H2S04 + 20N*SO,H Nit rosusul pliuric Ni troiosulphoiiic acid. acid. 20N.S03H + 20N*0*803H = 4N0 + 20(S03H) Pyrosul phuric acid. According to this two molecules of sulphur dioxide and on0 molecule of oxygen are absorbed simultaneously by the liquid mist particles and under the influence of the catalyser which is supposed t o be alter- nately nitrososulphuric acid and nitrososulphonic acid sulp tiuric acid is continuously produced.H. M. D. Electrolytic Preparation of Persulphates. CONSORTI UN F u i t Compare Muller and Priedberger Abstr. 1902 ii 450; and Levi Abstr. 1903 ii 474).-ln the preparation of potassium persulphate by electrolysis without a diaphragm the yield is greatly increased by the addition of hydrofluoric acid becoming equal to that obtained in the preparation of ammonium persulphate. Potassium persulphate may thus be prepared directly instead of by double decomposition from the ammonium salt. There is no loss of fluorine and the anode remains bright throughout the process.In the electrolysis of ammo- nium sulphate the yield of persulphate is not increased but no evolution of nitrogen is observed. ‘the influence of fluorine ions on electrolytic oxidation has been observed by Skirrow (Abstr. 1903 ii 69). C . H. D. ELEKTROCHEMISCHE INDUSTRIE t!k ERICH ?dULLE€L (D.R.-P. 155805.84 ABSTRACTS OF CHEMICAL PAPERS. Aution of Hydrogen Sulphide on Selenious Acid. I. Sensitiveness of Selenium Sulphide towards Light. ALEXANDER GUTBIER and J. LOHMANN (Zed. anorg. Chem. 1904 42 325-328).- When the yellow hydrosol obtained by passing hydrogen sulphide into an aqueous solution of selenious acid is boiled in presence of hydrochloric acid a bright red precipitate is formed. This formation of the red hydrogel may also be induced by the influence of light.From the experiments described it appears that the formation of red aelenium sulphide is conditioned by the four factors heat time light and pressure. A. McK. Reactions between Nitric Oxide and Oxygen or Atmo- spheric Air. GEORGLUNGE and E. BERL (Chem. Zed. 1904 28 1243-1245).-The formation of nitric acid by the action of an excess of oxygen on nitric oxide is chiefly dependent on the amount of water present. With an optimum of the latter nitric acid is formed in quantitative yield; with larger amounts of water nitraus acid is formed in addition and the more water is present the greater the amount of nitrous acid formed. The presence of free nitrogen when atmospheric air is substituted for oxygen has no appreciable effect on the reaction since nitric acid is also produced in this case in quantitative amount if the optimum quantity of water is present.The formation of nitric acid is a time- reaction. A. McK. A Method for the Preparation of a Mixture of Nitric Oxide and Nitric Peroxide. J. MATUSCHEK (Chem. Zeit. 1905 29 31)- A mixture of nitric oxide and nitric peroxide is Formed by the action of an aqueous solution of ferric chloride on sodium nitrite as represented by the equations Fe3C1 + GNaNO = Fe,(N0,)6 + 6NaCl and Fe,(NO,) + 3H,O = Fe,(OH) + 3N0 + 3N0. Ferrous sulphate or ferric sulphate may be substituted for ferric chloride in the reaction. If sodium nitrite is placed under carbon disulphide and hydrated ferric chloride suspended in carbon disulphide is added nitric oxide is alone evolved the nitric peroxide remaining dissolved in the carbon disulphide.A. McK. Theory of the Action of Metals on Nitric Acid. EDWARD DIVERS (J. Soc. Chew&. Ind. 1904 23 (24) 1182-1185. Compare Trans. 1883,43,443;Veley Abstr. 1891 i 525; 1892 i 41O).-Silver and mercury only act on nitric acid in the presence of nitrous acid which acts as catalyst and molecular quantities of nitrite and nitrate are primarily produced according to the equation 2 Ag + 2HN0 = AgNO + AgNO + H20. I n this interaction nitrous acid nitrogen per- oxide and nitric oxide only appear as secondary products. The nitrous acid is formed by the action of nitric acid on the primary nitrite and the nitrogen peroxide by the interaction of the nitrous and nitric acids when the proportion of water is small.Nitric oxide results from the decomposition of nitrous acid when the proportion of water is large and that of nitric acid small. The primary products of the action of zinc or tin on nitric acid areINORGANIC CHEMISTRY. 85 zinc or stannous nitrate and either ammonia nitrous oxide or nitrogen these three reduction products being produced independently of each other and without the formation of intermediate products. The primary formation of hydroxylarnine only takes place if a stable acid such as sulphuric or hydrochloric acid is present. Nitrous acid or nitrite only appears as a secondary product and nitric oxide and nitrogen peroxide are to be placed in the same category. The author thinks it possible to place all the metals in regard t o their action on nitric acid in one of the above two classes.Bismuth and copper belong to the former whilst all the strongly basic metals belong to the latter. H. M. I). Action of Carbon Dioxide on the Hydroxides and Carbon- ates of the Metals of the Alkalis and Alkaline Earths. PAUL N. RAIKOW (Chem. Zeit. 1904 28 1247-1252).-Normal potassium carbonate in aqueous solution is completely converted into potassium hydrogen carbonate by an excess of carbon dioxide no tetracarbonate being formed. The experiments made did not indicate the existence of the sesqnicarbonate K,H,( CO,),. When normal lithium carbonate is dissolved in water containing carbon dioxide lithium hydrogen carbonate is probably formed ; a little lithium tetracarbonate is probably also produced.By the action of water containing carbon dioxide on normal sodium carbonate the latter passes comple’ely into the sodium hydrogen salt the intermediate formation of sodium sesquicarbonate not having been observed. Similarly in the formation of riibidium hydrogen carbonate from the normal salt no tetracarbonate was formed ; calcium strontium and barium carbonates respectively under similar conditions did not form tetracarbonates. Evidence is submitted to indicate the formation of hydrogen carbonates of lithium calcium strontium and barium when an excess of carbon dioxide is passed into water containing the hydroxides or normal carbonates of those metals A. McK. Affinity of Alkali Oxides towards Various. Anhydrides. D. G. GERASSIMOFF (Zeit.nnorg. Chem. 1904 42 329-340).-The author has studied the action of sulphur trioxide on alkali tungstates and vanadstes respectively and the action of carbon dioxide on the latter salts and also on alkali niobates tantalates titanates and aluminates respectively. The avidity of alkali oxides towards carbon dioxide and sulphur trioxide respectively increases with the molecular weight of the alkali oxide. This is the case for all systems in which the following anhydrides take part CO and SiO (Wittorf) CO and TiO (Smith) CO and WO CO and V,O SO and WO SO and V20 (the author). By coqarison of the relative affinity of the non-volatile anhydrides it is found that in the systems where (a) sulphur trioxide and tungsten trioxide and ( b ) sulphur trioxide and vanadium pentoxide are respectively concerned vanadium pentoxide has a greater affinity than tungsten trioxide. When carbon dioxide competes with one of the anhydrides V205 Nb,O SO WO A1,0 TiO and Ta205 the VOL.LXXXVIII. ii. 786 ABSTRACTS OF CHEMICAL PAPERS. sequence quoted represents the relative affinity of the latter anhydrides The following indicates the number of molecules of carbon dioxide expelled by one’molecule of the various anhydrides from two molecules of sodium carbonate a t 880° where the partial pressure of the carbon dioxide is 0.07 atmospheres V,O 2.000 ; Nb,O 1.891 ; SiO 1.310 ; WO 1.047 j A1,0 1,019 ; TiO 0.779 ; Ta,O 0.72‘7. A. McK Dissociation of Ammonium Chloride in its Analytical Relations. LUIGI SANTI (Chem. cent^. 1904 ii 1625-1626; from Boll.Chint. Fawn. 43 673-681).-Iron drillings are readily attacked by a hot concentrated solution of ammonium chloride with evolution of hydrogen and ammonia and yield the compound FeC12,2NH,01. Magnesium powder is attacked even in the cold forming the double salt llgCI2,2NH,C1. Calcium zinc cerium manganese and tin are also more or less attacked. On heating cobalt and nickel oxides with dry ammonium chloride they are reduced to the metallic state but other oxides are converted into chlorides On boiling calcium nickel manganous and ferrous oxides with a solution of ammonium chloride they are converted into chlorides with evolution of ammonia ; sesqui- oxides such as ferric oxide are not attacked. Carbonates of calcium magnesium copper cobalt and nickel are also converted into chlorides with evolution of ammonium carbonate.The sulphides of tin and antimony are attacked by heating with dry ammonium chloride whilst the sulphide of manganese and in a less degree that of zinc is attacked even by solutions of ammonium chloride Sodium thio- sulphate solution on boiling with ammonium chloride is decomposed according to the equation Na,S,O -!- 2NH4C1 = 2NaCl + H,O + ZNH + SO + S ; if heated on a distilling apparatus the distillate contains ammonium sulphide and polysulphide. BIanganates are converted into permanganates chromates into dichromates ; the Iatter then even yield free chromic acid. Barium dioxide when treated with ammonium chloride in the cold yields hydrogen peroxide but on heating there is an abundant formation of oxygen.Potassium persulphate yields chlorine and nitrogen and even hypochlorous acid. Glucosides (salicin amygdalin) are practically unaffected by ammonium chloride. L. DE K. [Silver Dichromate.] OTTO MAPER (Bey. 1904 3’7 4646).-A reply t o Autenrieth’s claim for priority (Bw. 1904 37 3886). W. A. D. Certain Properties of the Alloys of Silver and Cadmium. T. KIRKE ROSE (Proc. Roy. Xoc. 1904 ’74 218-230).-The investi- gation consists in a determination of the temperatures of solidification of alloys of different composition and i n a study of their micro- structure. Evidence has been obtained of the existence of the COM- pounds AgCd? Ag,Cd AgCd Ag,Cc12 Ag,Cd and Ag,Cd. The solid alloys containing from 0-25 per cent. of silver consist of crystals of AgCd set in a matrix of cadmium Those witli 25-40 per cent.consist of the compound Ag,Cd set in a matrix which consists mainly of AgCcl,. The 50 per cent. alloy contains crystals of a substance richINORGANIC CHEMISTRY. 87 in silver set in a matrix consisting mainly of AgCd,. The matrix or eutectic solidifies a t 420° nearly 300' below the freezing point of tho crystals. The alloys containing 50-60 per cent. of silver consist above 420° of mixtures of two different solid solutions one composed chiefly of the compound AgCd and the other of Ag,Cd,. Traces of the eutectic freezing a t 420' are still visible. When more than 80 per cent. of silver is present the alloys consist of a mixture of two substances at temperatures between the liquidus and solidus curves but these unite to form a single solid solution a t points on the solidus curve.The alloys containing over 80 per cent. of silver do not ordinafily undergo segregation and are practically homogeneous and uniform in composition. They are well suited as a material for the manufacture of trial-plates. J. G'. P. Metallic Calcium. KURT ARNDT (Bey. 1904 37 4733-4738. Compare Abstr. 1903 ii 76).-Metallic calcium prepared on a large scale by the electrolysis of fused calcium chloride has been examined in detail by the author. The metal contains traces of silicon and aluminium as impurity and when acted on by water yields almost pure hydrogen. The sp. gr. is 1.54 but rises to 1.56 on remelting owing t o an increase in the percentage of silicon ; when purified by distillation the metal has a sp.gr. 1.52. Calcium melts a t 800' and sublimes even below this temperature; the vapour reacts very vigorously with atmospheric air and combines with oxygen and nitrogen even more readily than magnesium does. Calcium Carbide as an Explosive in Mining Operations. MARCEL P. S. GU~DRAS (Compt. rend. 1904 139 1225-1226).-An explosive mixture of acetylene and air obtained by the action of water on calcium carbide in a limited space and fired by means of an electric spark can be used for blasting operations in mining and the author describes a cartridge containing an air chamber and charged with calcium carbide (50 grams) and water separated by a membrane which can be mechanically pierced after the cartridge is placed in the bore-hole ; after five minutes the acetylene (15 litres) mixed with the air is fired electrically.Preparation of Barium. ANTOINE GUNTZ (Ann Chim Plqs. 1905 [ viii] 4 5-25).-A risunzS of work already published (com- pare Abstr. 1901 ii 385 ; 1902 ii 138 ; 1903 ii 369 410)-The purest specimen of the metal obtained contained 98.35 p e ~ cent. of barium melted at about 850' and boiled a t about 1150'. E. F. A. 31. A. W. M. A. W. Decomposition of Barium Nitrate by Heat. E. BASCH (Chem. Zeit. 1905 29 31).-The author corroborates Gottlieb's results (Abstr. 1904 ii 403). Decomposition of Alkaline Earth Carbonates by Alkali Chlorides in Presence of Water. H. CANTONI and G. GUGU~LIA (BzlZZ. Xoc. chim. 1905 [iii] 33,13-27. Compare Abstr. 1904 ii 334). -The iQfluence of temperature time and concentration on the rate of A.MoK. 7-288 ABSTRACTS OF CHEMICAL PAPERS. decomposition of barium carbonate by solutions of potassium and sodium chlorides has been investigated and some further observations have been made on the action of sodium potassium and ammonium chlorides on alkaline earth carbonates at the ordinary temperature over comparatively long periods. The results which are tabulated and graphically represented in the original show that under all conditions the action of potassitim chloride on barium carbonate is initially greater than that of sodium chloride and increases more rapidly with increase of concent ration rise of temperature and lapse of time although the last factor exerts less influence than the two former in differentiating the action of the two chlorides. A t the ordinary tempera- ture the activity of the three chlorides diminishes in the following order ammoniuni potassium sodium ; whilst the resistance of the three alkaline earth carbonates towards all three chlorides increases in the order barium calcium.strontium. SAutions of sodium or potassium chloride have practically. no action on strontium carbonate. T. A. H. Silicates. IV. EDUARD JORDIS and E. H. KANTER (Zeit. ano9.g. Chem. 1904 42 418-432. Compare Abstr. 1903 ii 475 542 595).-Details are given of the preparation and analysis of the silicates of the alkaline earths previously described. I n concentrated barium hydroxide solutions the metasilicate BaSiO,,H,O is formed. By the action of strontium hydroxide on metasilicic acid a compound was obtained colztaining 2/3Sr0 to 1Si0 ; another preparation contained the proportion 1 /3Sr0 1Si0,. By the action of calcium hydroxide on silicic acid the metasilicate was not formed but a compound containing 2/3Ca0 1Si0,.The reaction between silicic acid and calcium hydroxide in presence of 111 mol. calcium chloride solution was studied. Mixtures were obtained. The authors were unable to confirm the results of Wahl (Abstr. 1902 ii sol) who claims to hnve obtained the silicate BaSiO,,GH,O by the action of barium hydroxide on quartz. A. McK. Separation of Glucinum from Aluminium and Iron. G. VAN OORDT (D.R.-I?. 155466).-.The salts of glucinum with fatty acids dissolve in chloroform (Lacombe Abstr. 1902 ii 122). Glucinum may be separated from aluminium and iron by converting the mixed salts into acetates digehting with glacial acetic acid and extracting with chloroform in which the pure glucinum acetate only dissolves.The treatment with glacial acetic acid is necessary in order to convert the insoluble acetate into the crystalline form soluble in chloroform. A solution of glucinum hydroxide in hydrochloric acid is quite colourless when free from iron. C. H. D. Salts and Substances for Incandescent Lighting. HANS BUNTE (Chem. C'entr. 1904 ii 1627-1628 ; from J. Gasbel. 47 1011-1013).-Since Eitner has found that the temperature of the Bunsen flame is itself considerably higher (1390-1545") than that ofINORGANIC CHEMISTRY 89 tht? mantle at corresponding points (1 265- 1395*) the temperature of the latter cannot be appreciably affected by a catalytic action of the incandescent mass.According to Schmidt the light emitted by pure thorium oxide become; of a bluer t i n t as the temperature rises whilst the nature of the light from cerium oxide is the same a t all tem- peratures of the flame. When the proportion of cerium oxide to thorium oxide is gradually increased to 0.5 per cent. the lumino4ty of the mantle becomes p e a t e r and the light of a bluer tint. Further addition of cerium oxide up t o 1.5 per cent. still increases the luminosity but the light now appears to be tinged with red. The illuminating power is decreased by larger proportions of cerium oxide and the character of the light approaches more and more to that emitted by the pure oxide. The first effect of rise of temperahre is to increase the blue rays but when the light has attained a great intensity a11 the rays w e affected in prnctically the same WRY.It has been found that when mantles of different weights but of the Sam9 composition a r e employed the light obtained from the heavier mantles is not so blue as that emitted by the lighter. The high temperiiture of the cerium oxide in the fhme is partly due to ils state of division and partly t o the smnll amount present whilst the lighting effect is primarily the result of selective radiation. E. w-. w. The Element z6. 1'. fi. 1,~cog 1 ) ~ BoISBAUDRAN (Compt. rend. 1904 139 1015-1016).-Contrai~y to the statement of Urbain (compare thiq vol. ii 35) the author maintailis that the elemeiitai~y nature of 26 was fully establishetl by him in 1895 (compare Abstr.1896 ii 24'3) although he did not succeed in obtaining the element free from dysprosium. M. A. W. Basic Alumino-silicates containing Haloids. Z. WEYBERG (Centr. Min. 1904 729-794).-By fusing kaolin with calcium chloride Gorgeu (Abstr. 1888 228) obtained tetrahedra of t h e compound 3Si0.2,..7A120,,6C:i0,2Ca~l*. This result is confirmed and by fusing kaolin with calcium bromide tetrahedra with the composition 5Si0,,8A120,,1 2Ca0,4CaRr2 were obtained. In both cases prismatic crystals of another alumino-silicate were also formed. Kaolin when fused with a small amount of calcium bromide gave tetragons1 prisms with the cornposittion SiO A1,03,2Ca0. L. J. S. Products of Weathering of Silicates in Clay Volcanic and Laterite Soils respectively.JAKOB M. VAN REMMELEN (Zeit. anorg. Chem. 1904 42 265-314. Compare Abstr. 1902 ii i0).-In soils where the products of weathering contain much alumina relatively to silica all determinations of silica made by extracting it with hydro- chloric acid are without value since the extraction is incomplete. If however the soil after treatment with hydrochloric acid is agitated for several minutes with dilute sodium or potassium hydroxide a t about 50° the residual silicic acid is dissolved. The ratio of alumina to silica as determined in a large number of the products of weathering of alluvial plastic clays soils of volcanic origin and laterite soils respectively was not constant I n ordinary90 ABSTRACTS OF CHEMICAL PAPERS. alluvial clays the portion extracted from the disintegrated silicate by hydrochloric or by snlphuric acid indicated that the weathering had taken place in a single stage whilst with the volcanic and laterite soils examined various stages of weathering were recognisable the products consisting of a mixture of silicates. The progress of the weathering was best exhibited with the laterites the final product being in those cases hydrargillite.The metals of the alkalis and of the alkaline earths are also present in varying amounts in the products of weathering examined but not in sufficient amount to conclude that chemical compounds were present of the composition m(Si02),ra(A120,),o(lWO),p(H20) where m n 0 and p are whole numbers. The amount of those bases present diminishes the further the weathering has proceeded.Estimations of the amount of water in air-dried clays showed that ordinary clays contain very little water which is lost at 15' when the clay is exposed to an atmosphere dried with concentrated sulphuric acid. They contain about 2H29 which is lost at a higher temperature. The laterites examined contained more than 2H20. The more basic the products of weathering are the more readily do they dissolve in hydrochloric acid. The amount of iron oxide in various soils was also determined. This iron oxide generally contains a t the ordinary temperature a little more than 1H20 and a t 100' rather under lH,O. The constitution of the silicates obtained by the weathering of clays and the process of weathering itself are discussed. Tables of the various analyses made are appended.A. McK. Absorption of water by Clay. JAKOB M. VAN BEMMELEN (Zed. cworg. Chem. 1904 42 314-324).-Various soils were dried a t looo and the percentage of water absorbed at 15' under varying vapour pressures was estimated. The process of absorption is reversible ; the hydration and dehydration can be repeated as often as desired and the amount of hysteresis varies with the amount of absorption. Determinations of the rate at which the water absorbed was expelled show that it was not essentially different from the rate a t which water itself evaporates under similar conditions. Reduction of Manganese Oxides by Amorphous Boron. Preparation of a New Manganese Boride. RINET DU JASSONNEIX (Compt. rend. 1904 139 1209-121 l).-Manganese oxides are readily reduced when heated with boron in an electric furnace for a few seconds with a current of 400 amperes and 100 volts ; if excess of boron is used the product consists of a mixture of manganese and a new manganese boride MnB (compare Troost and Hauteville Abstr.1876 i 883) which can be isolat,ed in the form of a crystalline powder having a sp. gr. 6.2 at 15' by the limited action of chlorine on the mixture a t a dull red heat Manganese boride burns in fluorine at the ordinary temperature is attacked by chlorine or bromine a t a red heat whilst the action of iodine at a high temperature is only superficial; it forms a fusible borate when heated in oxygen is not attacked by nitrogen; it slowly decomposes cold water forming manganic hydroxide and boric Mid j A.McK.INORGANIC CEIEMI8TIiY. 91 dissolves in dilute hydrochloric acid and is attacked by hydrogen chIoride or hydrogen fluoride nitric or sulphuric acids; by the action of ammonia at lOOO' a compound is obtained which contains nitrogen is not attacked by acids and yields manganates with evolution of ammonia on fusion with alkali carbonates. 3%. A. W. Composition of the Four Sulphides of Manganese. JOHN C. O L S E N ~ ~ ~ W. S. RAPALJE(J. An2er. Chem. Soc. 1904,26,1615-1622). -An investigation of the sulphides of manganese has shown that three sulphides exist two of which the red and the green are anhy- drous whilst the grey sulphide contains a large proportion of water. The pink sulphide of manganese obtained by the action of ammonium sulphide on a qeutral solution of manganous chloride appears to be a mixture of the grey and red sulphides in varying proportions.The pink sulphide is not uniform in composition and contains from 4 to 14 per cent. of free sulphur and a varying amount of water. When heated a t 360° in an atmosphere of hydrogen it is gradually converted into the green modification. When precipitation is effected with colourless sodium sulphide a brick red sulphide of fairly constant composition is obtained contain- ing about 0.75 per cent. of water. On heating the pink sulphide with hydrogen sulphide the product obtained consists of lumps which are grey on one side and red on the other. The reason that the sulphide precipitated by sodium sulphide cannot be converted into the green modification whilst that produced by ammonium sulphide readily undergoes this change is probably that the latter product contains varying amounts of the grey sulphide.The green modification is the most stable form of the sulphide is more distinctly crystalline than the other forms and is probably more complex in structure. E. G. Influence of Water Vapour on the Reduction of the Oxides of Iron by Mixtures of Carbon Monoxide and Carbon Dioxide. OCTAVE BOUDOUARD (Compt. rend. 1906 140 40-42).- Comparative measurements of the reduction of ferric oxide by gaseous mixtures containing equal volumes of carbon monoxide and carbon dioxide have been made the gas being dried in one series of experi- ments and saturated with water vapour at room temperature in another. The gas was passed over the ferric oxide contained in a porcelain boat which was heated in a n electric resistance furnace maintained at; constant temperature each experiment lasting one hour.The following numbers give the percentage loss af weight of the ferric oxide Temperature ......... 400° 550° 800' 925O 1050' Dry gas ............... 0.87 4.3 4.0 5.6 6.5 Moist gas ............ 0.45 3.8 2.65 4.4 6.9 The reduction effected by the dry gas is considerably greater a t t,he lower temperatures but there appears t o ,be little difference in the action a t about 100OO. Similar data were obtained in experiments on92 ABSTRACTS OF CHEMICAL PAPERS. the reducing action of dry and moist carbon monoxide on ferrous oxide at 850". The author considers that the results are in harmony with the increased economy which accompanies the use of dry air in the blast furnace.H. M. D. Formation and Solubility of Double Chlorides of Iron and the Alkali Metals. F. WILLY HINRICHSEN and EUGEN SAcIisEL (Zeit. pAysikaZ. Chem. 1904 50 8 1-99).-Determinations of solu- bility and dilatometric investigation show that in the case of sodium and ferric chlorides there is no formation of double salt between 0" and 60'. At 21° potassium and ferric chlorides form a double salt of the composition FeC1,,2KC1,H20 ; if however the potassium chloride is in excess mixed crystals are formed (compare Itoozeboom Abstr. 1892 13S4). The temperature a t which the double salt is formed from the separate salts is 22.0-225O. No evidence could be found for the existence of the double salt FeC13,3KC1 referred t o by Werner.I n the case of cmiium and ferric chlorides a t 21° two double salts are stable (1) a yellow salt PeCl3,3CsC1,H,O produced when the solution contains excess of cresium chloride (2) a red salt FeCl,,2CsCl,H20 formed in presence of excess of ferric chloride a t 39.5-39-8'. The authors were unable to isolate the compound FeCI,,CsCl,&H,O described by Wslden (Abstr. 1895 ii 165). The compounds FeC1?,2KCI,H20 and FeC1,,2CsC1,H2O are in harmony with Werner's views regarding the 6' co-ordination number " (see Abstr. 1902 ii 554). The other msium double salt FeC1,,3CsCI,H20 is not in harmony with these views. Incidentally the solubility of cmium chloride has been determined at several temperatures between Oo and 40°. J. C. P.Preparation of Iron Pbosphide from Calcium Phosphate. GUSTAVE GIN (D.R.-P. 156087).-1ron phosphide may be prepared in a closed electric furnace from calcium phosphate and iron silica being added as a flux. Iron pyrites may be employed as a reducing agent instead of the carbon hitherto uked 4Ca,P,08 + 12Si0 + 5FeS2 + 1 1 Fe = 8Fe2P + 12CaSi0 + 1 OSO ; or 5Ca,P20 + l'5Si0 + 10 FeS + 5Fe,O = 10Fe,P + 15CaSi0 + 20S0,. The temperature required is slightly above the melting point of iron. Several arcs each with ft tension of 25 to 30 volts enter the same crucible and the energy consumed amounts to 50-60 volts per sy. cm. of electrode surface. Acid calcium phosphates may be em- ployed to economise flux. Some free phosphorus is formed by secondary reactions and is absorbed in a chamber containing iron borings.C. H. D. Alloys of Cobalt and Nickel. W. GUERTLER and GUSTAV TAMMANN (Zeit. anorg. Chenz. 1904,42 353-362).-The authors have studied the melting-point curve of alloys of cobalt and nickel to deter- mine whether a compound of these metals is formed or whether both metals separate completely or incompletely from the molten mass.INORGANIC CHEMIS'l'ltY. 93 When cobalt is heated i t is like nickel converted into a stable non- magnetic variety. I n the curve represented where the abscissz indicate the coinposition of the mixture and the ordinates the melting points observed the melting-point curve is represented by a straight line the course of which is expressed by the formula Ato = 0.35p where p indicates the percentage of cobalt and At the rise of melting point.Thevariation of each individual melting point from this straight line amounts to only + 5". Nickel and cobalt accordingly separate from their molten masses in mixed crystals. Cobalt is transformed into the non-magnetic variety at IlliO" whereas nickel is transformed at 333". The equilibrium curve of the magnetic and non-magnetic crystal forms of nickel cobalt and their alloys mas studied. The transition temperature of alloys containing 10 per cent. of cobalt rises from 100" to 60° from which the conclusion is drawn that the concentration of the single crops of mixed crystals which separate from the same molten mass cannot differ from one another by more than 2 to 3 per cent. Various mixtures of nickeland cobalt were taken and the lowest temperature determined a t which those mixtures became non-magnetic ; the temperatures were also noted a t which the mixtures again became magnetic after having been cooled.The melting point of gold is 1044'. I n the concentration-temperature diagram sketched in the paper the melting-point curve separates the field of the molten mass from the field of non-magnetic crystals whilst the transition curve separates the latter from the field of magnetic crystals. At the temperatures of the melting-point curve from the melting point of pure nickel 1484" to the melting point of pure cobalt 152S0 the composition of the molten mass and of the mixed crystals which separate is approximately identical. A t the temperatures of the transition curve the magnetic mixed crystals are in equilibrium with the non-msxgnetic mixed crystals.A. McK. Triamminecobalt Salts. A New Case of Hydrate Isomerism. ALFRED WERNER and ADOLF GRUN (Be?*. 1902 37 4700-4706).- CAZorodiaquotriccmmineco balt nitrate [ CoCl( N H3)J *NO,,. formed by the addition of a freshly-prepared solution of chlorodiaquotri- amminecobalt chloride to nitric acid of sp. gr. 1-4 separates in bluish- violet needles. The corresponding bromide CoCl(OH,),( NH3)SBr2 prepared by the addition of the chloride to hydrobromic acid of sp. gr. 1-49 cryatallises in dark blue needles. It is very unstable and in a moist atmosphere forms c ~ l o r o b r o r n o a p u o t r i m ~ i i ~ ~ e c o b a l t bromide [CoClBr(OH,)(NH,),!Br a green salt from which the original bromide may be regenerated in small amount by suspending it in alcohol and gradually adding water. The solution of chlorodiaquotriamminecobalt bromide in water is blue and when heated becomes red.When hydrobromic acid is added t o the blue solution the isomeric chZorobromoapuotriamn.LinecobaZt bromide [CoClBr( OH,)(N H,),]Br,H,O separates in brown needles even at temperatures below 0". It is more stable than the blue iso-94 ABSTRACTS OF CHEMICAL PAPERS. meride and is converted into the green chlorobromoaquotriammine- cobalt sulphate by the action of concentrated sulphuric acid. It may also be prepared by the action of hydrobromic acid on dichloro- aquotriamminecobalt chloride when it separates in glistening chocolate-coloured leaflets. Below 0' it forms a green solution with water at the ordinary temperature a bluish violet and at higher temperatures a red solution. When chlorodiaquotriainminccobalt sulphate is triturated with fuming hydrobromic acid anhydrous chlorobromoaquottriamminecobalt bromide separates in olive-green crystals.The structural relationship between the three bromides described is expressed as follows [clCo&2k]Br2 (blue bromide) Br,H20 (brown bromide) A. McK Existence of a Normal Green Chromic Sulphate. ALBERT COLSON (Compt. rend. 1905 140 42-44).-When the green solution obtained by reducing chromic acid by means of sulphur dioxide a t 0' is completely evaporated in a vacuum an amorphous green hygroscopic substance of the composition Cr2(S0,!,,10H,0 is obtained. The sub- stance appears to be a normal chromiuin sulphate; its aqueous solu- tions do not contain free sulphuric acid and the green colour of the solutions is permanent.On addition of barium chloride a considerable proportion of snlphuric acid is precipitated. It differs from the green sdphate described by Kecoura (Abstr. 1892 i 411) for the green colour of solutions of this soon changes to violet and barium sulphate is not precipitated on addition of barium chloride. On boiling t'he aqueous solution a change takes place which is repre- sented by the equation 2Cr,(S0,)3 + H20 = Cr,O(SO,) + H,SO,. The density of the solution is smaller after boiling and the heat development on addition of barium hydroxide indicates the presence of one molecule of free acid in the boiled solution for every two mole- cules of the original salt.The freezing point of the solution is un- altered by the change which takes place. The salt is supposed to have the constitut'ion SO,< I Cr:SO Cr:RO H. M. D. Uranyl Selenide and Potassium Chromic Selenide. JAR. MILBAUER (Zeit. unorg. Chem. 1904 42 450-452).-17ranyl selenide U02Se prepared by heating a mixture of selenium potassium cyanide and uranium oxide (prepared from uranyl acetate) a t a low red heat forms black hexagonal prisms with a metallic lustre and closely resembling uranyl sulphide. When cold hydrochloric acid is added to it hydrogen selenide is evolved and uranyl chloride produced. It is acted on very vigorously by nitric acid selenium a t first separating and then gradually oxidising.INORGANIC CHEMISTRY 95 Potassium chwmic seleizide K2Cr2Se4 prepared by heating a mixture of seleninm potassium cyanide and chromium sesquioxide a t a red heat forms dark green hexagonal crystals which are readily soluble in nitric acid but; insoluble in hydrochloric acid.A. McK. Action of Sodium Hyposulphite on Metallic Salts. 11. OTTO BRUNCK (Annalen 1904 336 281-298. Compare Abstr. 1903 ii 481).-On addition of a solution of sodium hyposulphite to solutions of metallic salts mostly one of three reactions takes place ( a ) the pre- cipitation of a sulphide ( b ) reduction to the salt of a lower oxide or ( c ) complete reduction to the metal. The following reactions with sodium hyposulphite are described. Thallium salts a t the ordinary temperature and indium salts on boiling yield the sulphides but incompletely owing to the action of sulphurous acid formed.I n a neutral solution stannous chloride forms a precipitate of stannous hyposulphite which dissolves in an excess of the reagent but in presence of an excess of acid stannous sulphide is formed ; solutions of stannic chloride are completely precipitated as the sulphide. I n neutral solutions lead salts are completely precipi- tated as the yellow hyposulphite which changes into the black sulphide. Molybdic acid forms the sulphide in neutral solutions whilst tungstic is reduced in slightly acid solution. Nickel and cobalt salts form the sulphides quantitatively in neutral ammoniacal or weak acetic acid solutions. Titanium tetrachloride is reduced in neutral solution to the tri- chloride but not in alkaline solution as the action is reversed; similarly chromates are reduced to chromium oxide permanganates to manganese salts ferric to ferrous salts and on heating to ferrous sulphide and platinic to plntinous chloride with precipitation of sulphur.Salts of palladium selenates and tellurates are reduced in neutral! solutions completely to palladium selenium and tellurium respectively. Similarly arsenic is obtained from its oxygen compounds as a brown powder which if the reduction takes place in a strongly acid solution contains arsenic trisulphide and sulphur. The behaviour of salts of antimony and of bismuth is similar to that of the salts of copper. Solutions of salts of germanium are apparently unaffected at the ordinary temperature but on warming sulphur is precipitated. G. Y. Inactive Thorium. CHARLES BASKERVILLE and FRITZ ZERBAN (J Amer. C h n . Xoc. 1904 26 1642-1644).-1t has been shown by Hofmann and Zerban (Abstr. 1903 ii 732) that inactive thorium can be obtained from certain minerals which do not contain any radioactive constituent. A new source of inactive thorium has been found in a rock from South America which is of a greyish-slate colour consists chiefly of barium carbonate containing a very small percentage of thorium and exhibits no radioactivity. E. G.96 ABSTRACTS O F CHEMICAL PAPERS. Decomposition of Antimony Hydride. ALFRED STOCK (Zeit pJLysiknZ. C'hem. 1904 50 11 1-1 12).-A continuation of a discussion with Bodenstein (see Bodenstein Abstr. 1904 ii 245 719 ; Stock and Guttmann ibid. 489). J. C. P. Purification of Tantalum. SIEMENS & HALSKE AKTIENGES. (D.11.-P. 155548).-Crude tantalun1,obtained by reduction with sodium always contains oxide which may be removed by fusion in a closed exhausted electric furnace. The crude tantalum compressed into a crucible of thoria or magnesia forms the anode; the cathode con- sists of a rod of pure tantalum or silver which is movable from the outside of the enclosing vessel. An arc is started and is then caused t o travel over the whole anode surface when fusion to tt homogen- eous non-porous mass occurs. C H. D.