ii. 60 ABSTRACTS OF CHEMICAL PAPERS. Inorganic Chemistry. Recovery of Perchlorate Residues &om Potassium Estimations. A. V~RTHEIM 9 (Chem. Weekblad 19 18 15 581-584. Compare A. 1917 ii 568).-The usual method for the recovery of perchloric acids is to transform all the reagent into its potassium salt which is easily purified by recrystallisatic and t o distil the mixture of this' salt with an excess of sulphuric acid in a vacuum. This method allows of recovery of only 25% of the quantity of acid originally used and moreover requires a good vacuum pump. A method has now been devised which consists of the addition of unslaked lime to the collected alcoholic filtrates filtration and washing of the precipitate with alcohol and recovery of the mixed perchlorates of calcium sodium and magnesium by distilling off the alcohol.The solid residue is warmed in a basin with sufficient 50% sodium carbonate t o convert the mixed salts completely into the sodium compound the whole filtered and the filtrate concentrated t o crystallisation point. A large excess of 40% hydrochloric acid is added the clear liquid decanted from the precipitated sodium chloride the last drops being separated on a vacuum filter and the solution of sodium perchlorate concentrated to D 1-125. A series of determinations carried out with recovered material gave uniformly higher results than duplicate tests carriedINORaANIC CHEMISTRY. ii. 61 out with pure perchloric acid. Investigation showed that the presence of sodium perchlorate in the recovered material diminishes the solubility of the potassium salt and that therefore more accurate results are obtained by using the recovered material than Appearance of Fogs in Chemical Reactions.VIKTOR ROTHNUND (Monntsh. 1918 39 571--601).-The nature of the fogs produced in certain chemical reactions has been investigated. I n the case of the fogs produced when ozone enters into reaction with a large number of reagents (particularly reducing agents) I t is shown tlhat the fog consists mainly of water in which a small quantity of the products of the reaction are dissolved. These fogs only Occur when the reducing agent is of a volatile nature and when the reaction products are soluble in water. The size of the fog particles is practically the same in a number of very different reactions.The approximate diameter as calculated from the rate of subsidence is 10-4 cm. The stability of the liquid drops in saturated water vapour is explained by the increased curvature brought about by the solution of the products of the reaction which effect a reduction of the vapour pressure. In the case of ozone and hydriodic acid a larger value is calculated for the diameter and this is explained by the formation of hydrogen peroxide. The ammonium chloride fogs the fogs from fuming acids fogs pro- duced by the action of radium emanatioii on sulphur carbon disulphide camphor and iodine as well as the electrically pro- duced and atmospheric fogs are also considered. In all cases the fog produced shows an analogous behaviour t o the ozone foes and generally ha! drops of about the same size.The Waters of the Atlantic Ocean on the Argentine Coasts. HERCULES CORTI and HECTOR H. ALVLREZ (Anal. SOC. Quim. Lilrgentiim 1918 6 108--120).-A detailed study of samples of sea-water taken under varying conditions a t different places on the Aiwntine coast. Extensive tabular staternen ts are given of chemical composition and of general and phvsico-chemical properties. The methods of Synthesis of Sufphuryl Chloride in Presence of Organic Compounds. GUIDO CUSXANO (Atti R. Accad. Lincei 1918 [v]. 27 ii 201-204).-The keto-ciueole described by the author an3 Linari (A* 1912 i 272) is similar in many of its chemical proper- ties to camnhor which it may replace in the preparation of sulphuryl chloride from sulphur dioxide and chlorine by Schulze’s method.The reaction SO? + CL = SO,C;1 is also accelerated by cyclohexanone. menthone or tetrahgdrocarvone but does not occur in presence of a-bromocamphor Reychler’s camphorsulphonic acid camphorquinone or rnonobromo- or monochloro-ketocineole (to be described elsewhere) with the halogen attached t o the methvlene contiguous to the carbonyl group; khe reaction is however by working with chemically pure perchloric acid. s. 1. L. J. F. S. analysis adopted are briefly indicated. w. s. M.ii. 62 ABSTRACTS OF CHEMICAL PAPERS. activated by introducing into the compounds just named a positive radicle for instance by transiorming the sixlphonic acid into either the amide or the anilide or camphwquinone into the monoxime or isonitrosocamphor. Pernitrosocamphor also catalyses this reaction but not the anilitlle of chlorocamphorsulphonic acid.These results are not in disagreement with the view that the catalytic action of camphor is connected with the formation a t low temperatures of one or more compounds of sulphur dioxide with camphor. T. E. P. New Method for the Synthesis of Ammonia. H. HAMPEL and R. STEINAU (Chen~. Zeit. 1918 42 594).-Metallic iron ammonium chloride and nitrogen are heated together a t 300° under 50 atmospheres pressure. The reaction proceeds according to the equation 3Fe + GNN,Cl+ 2N= 3FeC1 + 8NH,. A gaseous mixture is obtained containing up t o 99% of ammonia. The por- tion of this derived from the ammonium chloride may be allowed to react with the ferrous chloride; the ferrous hydroxide formed is reduced and the animonium chloride recovered for further use.[See further J . Soc. CAem. Im?. 1929 February.] W. P. S . Mixtures of Nitrogen Peroxide and Nitric Acid. TTILLIAM ROBERT ROUSFIELD (T. 1919 115 45-55). Oxidation Pressure Limits (A Theory of the Pressure Limit in Antoxidation). W. P. JORISSEN (Chem. Weekblnd 1918 15 705-714).-A r6sum6 of the observations and measure- ments on the lower limit of pressure in the autoxidation of phcw phorns in oxvgen and of the theories which have been advanced to account for the facts observed. The author considers the phenomenon to be a particular case of the ignition of an inflammable gas mixture (phosphorus vapour and oxygen) whenever the ratio of combustible constituent t o oxygen becomes sufficiently large the temperature in this case being sufficiently high to cause spontaneous ignition.W. S. M. The Atomic Structure of Carborundum determined by X-Rays. C. L. BURDICK and E. A. OWEN (J. Amer. Chem. Soc. 1918 40 1749-1759).-The angle between the edges of the elementary rhombohedron of carborundum which crystallises in the ditriqonal pyramidal class of the hexagonal system is 89O56*6’ the departure from the simple cube being due to a shortening by only 0*150/ of one trigonal axis of the cube. The results of the X-ray rneasiarements here recorded irsing rays from a palladium tarqet. show a very close relationship between the crvstal structure of carbornndum and t h a t of diamond. The silicon and carhon atoms in the carhoriindiim crvstal are each arrsnped on face- centred rhombohedra1 (nearlv cubic) lattices.In the mism Dlanes (lOT01 and (11%). the carhon and silicon atoms lie in the same planes from which i$ follows t h a t in the direction of the principalINORGANIC CHEMISTRY. ii. 63 axis the carbon and silicon atoms alternate there being no lateral displacement between the two kinds of atom with respect to this axis. From the relative intensities of the reflections of the spectra of different orders by the different planes the displacement of the carbon planes from the silicon planes in the direction of the prin- cipal axis is calculated to be equal t o 0-36 of the distance between two consecutive carbon o r silicoii planes. The carborundum Struc- ture can then he derived from that of the diamond by replacing the carbon atoms of one of the two iiiterpenetrating face-centred cubic lattices of the diamond by a similar lattice of silicoii atoms s?iorten- ing one trigonal axis by 0*15% and displacing the atoms of one lattice from a position 0.25 to one 0-36 of the distance between successive planes of t'he other lattice in the direction of the shortened axis.From the values Q€ the distances between the atom planes derived from the inea surements t-he demity of carborundum is calculated t o be 3-11 the observed value being 3.123. F,. H. R. Electrolysis of Potassium Phosphate. A. RIUS IT M I R ~ ( A rial. Fis. Q ~ i r n . 2918 16 573-6lQ).--The electrolytic oxida- tion of potassium phosphates is assumed to1 take place in two stages the first. being the formation of perphosphate and the second that of monoperphosphate.F o r dipotassirim hydrogen phosphate the primary reaction can be represented by the equation 2K2T-TB0 + 0 = K.%P208 + E,O. The perphosphate then undergoes fhrther oxidation to monoperphosphate H,P,O -+ 0 + H20 = 2K,RPOi. A. J. W. The Fusion of Sodium Hydroxide with some Inorganic Salts. ~IAITLAXD C. HOSWELL and J. V. DICKSOY (J. Bnzer. C7mnz. SOC. 1918 40 1773--1779).-Whea certain salts which can function as osygen acceptors are fused a t high temperatures (30O-4QO0) with sodium hydroxide trhe salt is oxidised and hydrogen evolved. Quantitative experiments have been made with sodium arsenite and ferrous sulphate and it is found that the hydropa given off is equivalent t o the oxygen taken up by the salt. Stannous chloride and vanadium srilphate are also oxidised and cerotis and uranous sulphates to a small extent.It is remark- able that such readily oxidisable salts as sodium nitrite and sodium sulphite are not oxidised when fused with sodium hydroxide. It i s perhaps noteworthr thatl all the bases the salts of which were foiind t o be oxidieed belong with the exception of iron to groups 4. 5. or 6 of the periodic table The mechanism of the reaction consists ultimatelv in the decom- position of water. the oxygen carrying the oxygen acceptor to a higher staqe of oxidation whilst the hvdrocen is evolved as gas. The general oxidisiiig action of water catalvsed bv sodium hvdr- oxide is also shown by the evolution of hydrocen when such metals as zinc and aliiminium are boiled with sodium hydroxide solution.33. H. R.ii. 64 ABSTRACTS OP CHEMICAL PAPERS. Some Properties of Magnesium Ammonium Phosphate and Magnesium Pyrophosphate. %. KARAOGLANOV and P. DIMITROV (Zeitsch. anal. Chem . 1918 57 353-371).-The con- version of magnesium ammonium phosphate into magnesium pyro- phosphate by ignition somet.imes is and sometimes is not accom- panied by incandescence. The presence or absence of the pheno- menon is found to depend on the coiiditions under which the mag- nesium ammonium phosphate is precipitated. If precipitated slowly a t the boiling temperature the productl does not incandesce but if formed quickly a t lower temperatures it invariably does whether precipitated from a magnesium solution o r from a phos- phoric acid solution. The pyrophosphate formed with incandes- cence is grey t o black in colour.whilst that formed without in- candescence is quite white. The authors conclude from their experiments that only samples of magnesium ammonium phosphate which contain traces of organic matter will incandesce on ignition. since when organic matter such as filter paper is carefully ex- cluded no incandescence is observed. There are apparently two modifications of magnesium pyrophospfiate differing considerably in physical properties. The one formed without incandescence is loose i n texture and white in colour; the other the formation of which is always accompanied by incandescence is hard and lava- like grey to black in coloixr and more resistant to hydrochloric and nitric acids. Its colour is due to enclosed particles of carbon and 4s only with difficulty removed at; a very high temperature by ignition but can be destroyed hy treatment with acids or an oxidising agent such as ammonium nitrate.The incandescence of any sample of magnesium ammonium phosphate can be prevented by evaporating it before ignition with an ammonium salt or by heating very slowly t o the decomposition temperature. The opinion of Balareff (A. 1917 ii 90) that the properties of the magnesium uyrophosphate obtained depend on the vapour tension (degree of hydration) of t'he rnagiiesimm ain~oninmn phosphate before calciii- ation is shown t o he incorrect. E. H. R. Adsorption of Metals from Drinking Water by Glass. K. SCRERINGA (Ph,arrn. TVeekblad 1919 56 8-9).-The propor- tion of lead in drinking-water can considerably diminish within a few hours.T t is known that if an aqueous magenta solution be boiled in a glass vessel the latter cannot afterwards be cleaned in the ordinary way; this absorption however does not occur i f the vessel has been previously cleaned carefully with soap and water and afterwards well rinsed out. Since then organic dyes are not adsorbed bv a cleaned glass stirface it anpears very doubtful that metallic salts should be so adsorbed. This conclusion was con- firmed bv estimating colorimetrically soltitions of various salts which had been allowed to remain f o r two days in carefiilly cleaned glass vessels. I n no case was the sliqhtest diminution of &he amount of metal in solution detected. It appears therefore that the diminution in the case of lelad is due t o chemical action eitherINORGANIC CHEMISTRY.ii. 65 by disturbance of an equilibrium when the water is removed from contact with the lead or by precipitation of finely divided lead carbon ate. s. I. L. Anhydrous Mercuric Fluoride. OTTO KUFF and GUSTAV BAHLAW (Ber. 1918 51 1752-1760).-;rLrrzhyd’?.o.us mercuric fiuorzde may be prepared by heating mercurous fluoride i n a current of dry chlorine a t 275O o r of dry bromine at 400° o r by heating mercurous fluoride at 4503 under 10 mm. pressure. Mercuric fluoride forms transparent octahedral crystals m. p. 645”; its b. p. is estimated a t 65W. *4ttempts to determine the vapour tension at; various temperatures did not yield satisfactory results since the vessels are attacked by the vapours.The sub- stance is very sensitive to moisture and becomes discoloured by traces of water vapour which are not analytically demonstrable; on exposure to air hydrogen fluoride is evolved and mercuric oxyfiuoride and ultimately mercuric oxide remain. With small quantities of water a white hydrated oxyfluoride is formed whilst with larger quantities mercuric oxide is gradu- ally produced. Mercuric fluoride dissolves in hydrofluoric acid solution (40%) and on cautious evaporation the hydrated fluoride NgF,,ZH,O is obtained in small colourless crystals. The vapours of mercuric fluoride attack platinum above 500°; mixttures of the fluoride with silver copper lead aluminium magnesium zinc tin chromium iron or arsenic react vigorously when strongly heated locally yielding amalgams and metallic fluorides tThe latter being easily isolable in the pure condition i f an excess of mercuric fluoride is used.Sulphur tetrafluoride appears t o be formed when mercuric fluoride is heated with sulphur but no reaction occurs with amorphous or graphitic carbon. The fluorine does not appear to be replaced when the fluoride is heated in a stream of chlorine ar bromine. Merczric chZoro&uoride HgClF is obtained as a pale yellow substance by passing dry chlorine over mercurous fluoride at 120°; the pale yellow bronzo$uo~ide is similarly prepared a t 105O. The preparation of mercuric flnoride in quantity is best effected by the p~ocess first described. It has D1j 8.95. Hg3E’,(OH),,3H,O H. W. Double Catalytic Process in the Oxidation of Aluminium in the Presence of Mercury.Oxidation of Aluminium Powder at the Ordinary Temperature. P. RONCERAY (BULL Sci. Pharmacol. 1918 25 193-198; from Chem. Zentr. 1918 ii 699).-AIuminium in a fine state of division is oxidised by t.he air and under watei. Pieces of aluminium in contact with iron do not oxidise in water but in the prepence of a small quantity of mercury aluminium tindergoes oxidation through the operation of two catalytic processes. The mercury owing to the formation of an amalgam reduces the aluminium t o a molecular state and the mercury oxide produced acts as an agent for the transfer ofii. 66 ABSTRACTS OF CHEMICAL PAPERS. oxygen from the air to the aluminium. Pieces of aluminium after rubbing with mercury oxide are rapidly osiciised in the air and under water.H. 177. B. Solubility of Aluminium Hydroxide. E. H. ARCHIBALD and Y. EABASIAN (Tmns. R o y . Soc. Co~nuda 1917-1918 [iii] 11 l-B).-The solubility of aluniinium hydroxide in ammonia solu- tions of different concentrations and in similar solutions t o which various amounts of animoiiiurn or potassium nitrate had been added has been determined a t 20° and 3Qo. The method consisted in shaking the freshly precipitated hydroxide in sealed tubes with the solvent for twelve hours in it thermostat filtering evaporating msasured weights of the solution in a platinum crucrble igniting and weighing. The following weights of Al(O€€) dissolve in 100 C.C. of solution of ammonia a t 20° Normality of NH,OPP 0-050 0.100 0,125 0.200 0.500 1.00.Grams of id( 0*0070 0.0080 0.0250 0.0380 0.0450 0.0240. The solubility increases theref ore with the ammonia concentra- tion t o a maximum which lies atl 0*5i’\’-ammonia and then decreases. In these experiments it is shown that the amountl of aluminium hydroxide in solution after thirty minutes’ shaking is considerably more than the above quantity but decreases as the shaking pro- ceeds until equilibrium is reached which is always achieved in less than twelve hours. The presence of ammonium nitrate de- creases the solubility of aluminium hydroxide in ammonia and the decrease is greater the higher the temperature and the concentra- tion of ammonium nitrate. The addition of potassium nitrate increases the solubility sf aluminium hydroxide to a very marked extent. It is pointed o u t that in the quantitative estimation of aluminium only a small excess of ammonia should be used and a 10% solution of ammonium nitrate should be used in the washing.J. F. S. Relation between Molecular Structure and the Activity towards Hydrogen Sulphide of Oxide of Iron. G. WEYMAN ( J . Soz. Chertt. Ind. 1918 37 333-336~).-Iron oxides obtained by heating a t looo to 650” are equally active as regards absorption of hydrogen sulphide in the cold but at 750° a change occurs which may also be effected a t lower temperatlures by very pro- longed heating. It seems t.hat the activity of the oxide is dependent primarily on molecular structure and not on any par- ticular degree of hydration but the oxide is derived in almost all cases from some form of hydrate.w. P. s. Chro niatscobaltiamrnines . SANUEL HENRP CLIFFORD BRIUGS (T. 1919 1’15 67-76). The Evolution and Oxidation of Chromic Hydroxide in Alkaline Solution. F. BOURION and A. S h h r A L (Compt. rend. 1919 168 59-62) .-An alkaline solution of chromic hydroxideINORGANIC CHEMISTRY. ii. 67 undergoes a change an keeping which tends to make it lose its chemical activity particularly in respect to its powers of reducing hydrogen peroxide. This change is the more rapid as the coil- centration of the chromium is greater and of the alkali is smaller. It is however possible to oxidise 97"/; of the chromium in chrome alum to chromate by adding sodium hydroxide t o the solut.ion of the alum containing four times the calculated quantity of hydrogen peroxide and then immediately destroying the excess of peroxide by shaking the solution with manganese dioxide.W. G. The Crystalline Structure of Grey Tin. A. J. BIJL and N. H. KOLKMEIJER (Chent. Weekblad 1918 15 1264).-A pre- liminary note on the crystalline structure of grey tin. The authors have shown that the crystals belong to the regular system. w. s. M. Bismuth Hydride and Polonium Hydride. FRITZ YAXETH (Bey. 1918 51 1704-1728).-By application of the methods used in the study of radioactive substances the author has succeeded in demonstrating that bismuth forms a gaseous hydride which possesses considerable stability a t the ordinary temperature and is not decomposed with much greater readiness than antimony hydride ; with increasing temperature the substance rapidly becomes less stable and is decomposed into its elements at a red heat.The gas can be almost completely condensed by the use of liquid air and subsequently in part regasified. Bismuth hydride is obtained by the solut-ion of an alloy of magnesium with thorium-C or radium4 in O*ZA'-hydrochloric or sulphuric acid. The alloy is prepared by exposing magnesium foil to the radiations of a radiothorium preparation contained in B glass capsule covered with silk paper which is impermeable to tihorium-X; shortly after its remo'val in consequence of the rapid fiecay of thorium emanation and thorium-A the deposit! consists 3ntirely of thorium-B and thorium-I:. The alloy is placed in a weighing bottle connected with an electroscope in such a manner ,hat a regular current of nitrogen can be sent through the tpparatus.After determination of the natural leak of the electro- icope O.2N-hydrochloric acid is dropped on to thel alloy; the !lectroscope. soon indicates an activity which becomes feebler after L few minutes. The results of this and similar experiments show when magnesium superficially alloyed with bismuth and lead s dissolved in dilute hydrochloric acid a few thousandths of the tismuth are converted into such a state that. they can be carried by a gas current through a cotton wool filter and that a similar eaction does not occur with lead. A series of controi experiments hows that the observed effects are actually due to a volatile cam- ound of bismuth and not for example to the liberation of horium-17 t o the selective action of the filter .or t o the relatively reater volatility of thorium-C chloride.The alloy Polonium hydride is prepared in a similar manner.ii. 68 ABSTRACTS OF CHEMICAL PAYERS. of magnesium and polonium is prepared by the electrolysis af a feebly acid solution of polonium chloride a piece of magnesium foil being used as cathode. When the alloy is dissolved i n dilute acid and the gases evolved are led into an electroscope an activation is observed which does notl decrease-in the course of a day and therefore is caused by polonium. The gas closely resembles bismuth hydride. It is remarkable that a current of oxygen can be used instead of hydrogen or nitrogen for t-he transport of the gas with- out sensibly diminishing the yield. The latter is less than in t,he case of bismuth hydride and up t o the present it has not been found possible to convert more than a few tenths’ part per thousand of tahe polonium into the gaseous state.If the gas current is cooled to -&Lo the polonium hydride is only partly condensed. The investigation of the gas is rendered very tedious since in each experiment the electroscope becomes permanently damaged and does not recover when left to itself for a time. H. W. Bismuth Hydride 11. FRITZ PANETH aad EBICH WINTERNITZ (Bey. 1918 51 1738-1743).-The application of radioactive methods having shown that bismuth hydride is capable of exist- ence and having indicated its mode of preparation and general properties (preceding abstract) the authors now describe attempts to prepare it in weighable quantity from non-radioactive material. Yhe requisite bismuth-magnesium alloy is prepared by heating equal weights of powdered bismut.h and magnesium (as free from silicon as possible) in an iron crucible i n a rapid stream oE dry hydrogen.The alloy is dissolved in approximately 4N-hydro- chloric or sulphuric acid (or in some cases nitric acid). Bismuth hydride is thus obtained in sufficient quantity to permit its detec- tion either by the formation of a bi$mut;h mirror or by luminescence tests. The bismuth mirror is obtained in the usual Marsh’s apparatus and very closely resembles t-he antimony mirror- As generally obtained it consists of a strong brown ring in front of and a fainter ring behind the heated spot. The former deposit appears to be frequently burnt into the glass and to be unsuitable for furbher experiments.This drawback can be overcome by placing a pierced clay disk on the tube and allowing t-he flame t o play against this as also by increasing the velocity of the gas current. The antimony arsenic and bismuth mirrors are most readily dis- tinguished by a number of chemical tests involving the use of sodium hypochlorite yellow ammonium sulphide hydrogen sulphide etc. which are fully described in the original. Attempts to estimate the yield of bismuth hydride by weighing the bismuth mirrors show that about 5 x 10-5 of the bismuth used is converted into the hydride or that the yield is only about one-twentieth of that obtained from thorium*. It should be noted however in this cannexion that circumstances have prevented the aukhors from determining the optimum conditions of experimenting.The authors have also applied Donau’s luminescence method (A,,INORUA31C CHEMISTRY. ii. fir! 1913 ii 743) to the ctetectioii of traces of bjsnluth aiid find the procedure very rapid and so sensitive that i t is rapable of proving tlis preseuce of traces of bismuth which cannot lie detected by the ring test. For this purpose the gases issuing from the Marsh's apparatus are ignited and a piece of pure calciunt carlwiiate is held 011 a plat,inum loop in the flame; the bismuth hydride i h clecoiiiposed and a poi-tion of the bismuth is deposited 011 the lime. The latter is alloxed to cool and then placed at the edge of the hydrogen flame wheii the preseiice of bismut,h is betrayed by the cornflower-blue luminesceme; in similar c ~ ~ - c u ~ ~ s ~ ~ I I c ~ s mtirnony is readily detected by a sky-blue lumiiiescence.Itot,h colora t.ioii5 being readily visible in bright daylight. The absorption of bismuth hydride by various reagents has bseii euamiiiecl ; t.he most suitable solutioii for this purpose appears Lo lie 0'4T-silver nitrate solution. Water absorbs the gas to some degree and 4AT-sulphuvic acid to about the same extent. O*S.T-Soc'iuin carbonate solution aiid -7'-potassium hydroxide 501~- tioii are more active whilst the gas is also absorbed by desiccating agents such as caleirim chloride or soda-lime. It is completely decomposed by concentrated sulphuric acid. An aqueous solution of hydrogen siilphide is not niore efficient than pure water.TI. w. Gold Amalgams. 3. PARRAVASO (Atti I t . Accud. L ~ ~ L c c ~ 1918 [v] 27 ii 168- -1'70; GoTxtta 1918 48 ii 123-128).- - Objections are raised to the argumeiit of Guertler (2ifetallogrophir 524) who concludes that mercury dissolves in gold giving origin t o a solid solution containing at inost about 10% of gold and that in the amalgams containiiig 90-Oo/ of gold no other crystalline individual is formed. The author has made the following experi- ments ( I ) Definite quantities of saturated *amalgam aiid of gold are placed in a vessel and the latter exhausted; distillation of the mercury on t o the gold then proceeds until the composit.ion of the amalgam formed reaches that of the solid in equilibrium with the liquid amalgam. (2) A current of hvdrogen is passed over the amalgam the quantity of mercury thus trailsported will be a fuiictioii of the t,eiision of the mercury of the amalgam provided that in any series of experiments the form of the apparatus and f lie velocity and duration of the gaseons current are maintained constant.The resuIts show that gold amalgams contain a t least the two compoiinds An,€€g? and Aii;Flg. T. FT. P. Amalg&ns. I. Colloidal Gold Amalgam. C. PAAL :mtl HERMANN XTEI'CTL (1<07loid Zcit,~ch . 1918 23 145- -3 58).-Hydro- sole. of gold amalgam have been prepared (1) by shaking gold sols with metaIlic mercury. (2) by allowing a gold sol t o remain in con- tact wit% metalIic mercury a t rest (3) bv mixing solutions of gold hydrosol with merciiry hy-drosol and (4) by precipitating both gold and inercixv from a solrition of the mixed chlorides i n the presenct. of a protecting colloid.I n the last-named case amalgam 'hydrosols VOL. cxvr. ii 3i. 7 0 ABSTRACTS OB CHEMICAL 1'Al'EES. of the coinpositioii represented by the €orniulze A~i,€lg aiid Au,tIg were prepared by reducing a iiiixtiir-e of suit'able concentration of the two chlorides in alkaline solution by hydraziiie in the presence of the sodinin salts of protalbic aiicl lysalbic acidh. The solutions after dialysis were evaporated to dryness ill a vacuum a i d gave brittle substances which possessed a black or greenish-black colour and a. ruetallic lustre. These substances readily dissolved in water with the production of colloidal solutiotis. The stability coin- positioii and general properties of the hydrosols varied with the niethod of preparation.J. F. S. Solubilities of Ammonium Platinichloride Platini- bromide and Iridichloride and the Separation of Platinum and Iridium. E. H. ARCHIBALD a d JOHN W. KERX (Trans. K o y . iSoc. ~ ' ~ i / ~ ~ l u 1917-- 1918 [iii] 11 '7- -16).-The solubility of the aiiirrioniuni salts of chloroplatinic bronioplatinic aiicl chloro- iridie acids has beeii cletermitied a t a series of temperatures in water aiid in solutions of various strengths of amnionium chloride and aniinoiiurn bromide. It is shown that 100 grams of water dis- solve 0.2902 gram of aiiiiiioiiiurri platiiiichloride a t 0*lo 0.3653 grain at 7 . 2 O . 0.4869 grarii a t 18'0° 0-5761) gram a t 25*4O 0.6370 gram a t 2 9 * 9 O 0.7870 gram a t 3 8 * 9 O 1.0131 graiiis at 4 9 .7 O . 1.4740 grams a t GO.2O l*744ff granis a t 70*0° 2.1800 grams a t $ 0 * 2 O 2.6150 grams a t 90.0° and 3.2515 g r a m a t 99O. AinmoniLuri platiiiibroinicle is somewhat rtiore soluble in water than the fore- going salt; 100 grams of water dirsolve 0.4165 gram a t O*:'O 0.5002 graizi a t 7.3O. 0.6438 orair1 a t 19.0° 0.7384 gram a t 25*0" 0.8147 gram a t %.To 1.03& granis a t 40*0" 1.2087 grains a t ZOO 1.5780 grarits a t 60*0° 1.9'365 grams a t 70-O0 2.3002 grams a t 80.0° 2.8370 grams a t 9O0 and 3,5866 grams a t 9 9 O . Arnrrioiiittm iridiuhloride is much more solriltle than either of the platinurn coiiipounds ; 100 grariis of water dissolve 0.5661 firmi at 0*2O 0.7055 grain a t 10*Oo. 1.0910 grams a t 25.O0 1-2066 grams at 30*Oo 1.5665 grams a t 40.0° 1.9664 grain.a t 50*Oo 3.4567 qarris a t 60-Oo and 4.3815 grarris a t 80*0*. In the presence of arii~iioiiiurti chloride the solubility of the ammoiiiuin salts of chloro- platinic acid aiid chloroiridic acid is much reduced but that of the iiyidichloride is several times as large as that of the platinichloride. Ammonium bromide reduces the solubility of the platinibroinide. In all three cases the reduction in the solubility is proportioiiaI t o the concentration of the ammoilium haloid. The differeuce in the solubility of aniinoniuiti platiiiicWoride arid iridichloride furnishes a good method for the complete separation of plat inurn and iridium. Ammotiiiim platiiiichloride i? appreciably less soluble than 1 )o t a ssi u m plat i 11 ichl or i d e . J. F. 8. Dehydrogenation of Palladium Hydrogen Hydrosol by Metallic and Colloidal Mercury. C. PAAL ttntl HERMANN STEYEB (Her. 1918 51 1743--1752).-Tt has beeii previously shown (Paal and Nartmanri. A . 1918 ii 303) that palladiumMIPr’ERALOGICAL OHEMISTRY. ii. 71 hydrosol gradually loses its catalytic activity in the presence of irtetallic or colloidal mercury; in the course of experiment,s on the action of the hydrosol on mixtures of hydrogen and oxygen in the presence of mercury it was observed that the catalyst speedily hecame passive and thatq the gradual slight absorption of hydrogen was preceded by a temporary increase in the volume of the latter. ‘Phis phenomenon forms the subject of the present communication i f 1 which i t is demonstrated that the palladium hydrogen hydrosol i.; decomposed by mercury with evolution of hydrogen and that tlehydrogenation is effected more rapidly by metallic than by volloidal mercury. Analysis of the residual palladium hydrosol and of the mercury proves that a portion of the palladium has passed into the mercury and that some of the latter has passed into the hydrosol. The action of colloidal mercury on palladium hydrogen hydrosol only leads to uniform results when care is taken that only the least possible excess of hydrazino is used as reducing agent ill the formation of the mercury hydrosol. H. W.