106 ABST tiACTS OF CHEMICAL PAPERS. In o r g a n i c C h e m i s t r y. Action of Hypochlorous Anhydride on Iodine Trichloride. Ry H. BASSETT and E. FIELDING (Chem. News, 54, 205--206).--Iodic anhydride is the main product of the reaction when solutions of iodine trichloride and’ hypochlorous anhydride in carbon tetrachloride are mixed, or when a current of bypochlorous anhydride is passed through a solution of iodine trichloride, or even over solid trichloride. In the last case, the action is very slow. D. A. L. Saturation of Selenious Acid by Bases. By C. BLAREZ (Conzpt. rend., 103, 804- 806). -With cochineal or helianthin (methyl- orange), selenious acid is monobasic. With litmus, it is monobasic to ammonia, lime, strontia, and baryta, but if soda or potash is used the litmus only becomes blue-violet when about 1.5 equivalent of alkali is added. With phendphthalein.and potash, soda, ammonia, lime or strontia, the colour change takes place when somexhnt more than 1.5 equivalent of alkali is added, but with phenolphthalein and baryta selenious acid is bihasic,, and the colour change is sharp and distinct. Selenious acid can be accurately titrated by means of standard barytn solution, using helianthin or phenolphthalein as indicator. The bavyta has double the value with the first indicator that i t has with the second. Both indicators can be used in the same solution ; the rose colour of the heliauthin disappears when the acid is half saturated, and the rose colour due to the change of the phenol- phthalein becomes visible when saturation ik complete.As no. basic barium selenite is formed when baryta is present in excess, an excess of baryta can be determined by means of phenol- phthaleh and a standard acid in the liquid containing the precipitated barium selenite. Selenious acid, like sulphurms acid, cant be estimated in presence of other acids by means of baryba, provided that the tofal basicity of these acids is indicated by helianthin. Helianthin and phenol- phthaleln are both a;dded to the same solution, and the amount of selenious acid is calculated from the quantity of bnryta solution required to produce the second colour change, this quantity being exactly half that which would be required to neutralise the selenious acid alone with phenolphthale’in as the indieator. C. H, B.Formation of Nitrites- By S. KAPPEL (Arch. Pkarm. [3], 24, 897--900).--The author has extended his observations made on the action of copper, iron, and zinc in contact with the air, and solutions of ammonia and the fixed alkalis (AbsBr., ’1883, 282, 286), employing in recent experiments magnesium, aluminium, and tin. Magnesium exposed to the a i r in contact with aqueous potash, gave small quan- tities of nitrous acid, ozone, and hydrogen peroxide. With ammonia solution, strong indications of nitrous acid were obtained. Alnmiriium in potassium hydrate produced nitrous acid readily, even in the cold,INORGAKIC CHEMISTRY. 107 nitric acid and hydrogen peroxide were also easily detected. With ammonia the action was much slower. Tinfoil gave scarcely any reaction ; nitrous acid was not produced, but hydrogen peroxide was perceptible.J. T. Compounds of Arsenious Anhydride with HaTogen Salts. By F. R~UORFF (Ber., 19,2668--2679).-1n this paper the preparation and properties of compounds of the halogen salts of the alkali metals with arsenious anhydride are described. These are best obtained by passing carbonic anhydride info a mixed solution of potassium arsenite and the halogen salt. These componnds separate either in the amor- phous form or in indistinct crystals, of the general formula MX,2As203, sparingly soluble in water, insoluble in alkaline carbonates, but very soluble in the free alkalis; when heated, they decompose with elimi- nation of arsenious anhydride, the compound KI, 2As,03 decomposing a t 350" ; KBr,2As20z a t 380" ; and KCI,As203 a t 240" ; a compound, KC1,2Asz0,, is also described.The ammonium compounds NH41,AsL03, NH4Br,2As,C3, and NH4Cl,As20,, are also described, those of sodium being reserved for a future communication. V. H. V. Some Probable New Elements. By A. PRINGLE (Chem. News, 54, 167-168).-TThe author claims to have discovered some new sub- stances, including five metals and a substance resembling selenium, called hesperisiztm, in some " Lower Silurian " rocks, situated in the county of Selkirk. One metal is said to be like iron, but gives neither the thiocyanate nor the tannic acid reaction ; one is Eke lead in appear- ance, is easily fused and volatilised, and yields yellow and green salts ; another which is charcoal-black, is called erebodium,; its equivalent is 95.4, and it forms several oxides.A fourth, g a d e n i m , with equivalent about 43.6, a light-grey powder, forms a red monoxide and a cream- coloured dioxide, yielding respectively white and yellow saltp. Another, polymnesturn (Pm), is a rather dark-coloured metal, with equivalent about 74. A preliminary description of four oxides, PmO, PmOs, PmO,, and (?)I Pm05, of two sulphides, PmS and PmSp, and of other compounds is given. D. A. L. Production of Alkali Metals. By H. Y. CASTXER (Chem. News, 54, 218--SlS).--Irou reduced by hydrogen or carbonic oxide is mixed with t a r in proper proportions, so that after the mixture is coked it has a composition about = FeCp The coke is finely ground, mixed with c a s t i c soda (or potash) in proportions = 3NaHO + E'eC,, or about 100 NaHO to 15 of coke.This mixture is introduced into a cast-iron crucible, and heated in a specially arranged furnace (described in the paper). The reduction and distillation commence at a temperatnre of 1000". When the operation is finished the crucible is removed, and another immediately put in its place. The residue consists of some sodium carbonate, and finely divided iron; the sodium carbonate is recovered, the iron used to make fresh reducing coke, aiid the crucible is used over and over again. The advantages over the old method are manifest. D. A. L.108 ABSTRACTS OF CHEMICAL PAPERS. Crystalline Scale formed in the Manufacture of Sodium Hydrogen Carbonate. By G. W. LKIGHTON (Amer. .J. Xci., 33, 318--.319).-This scale was formed in the manufacture of sodium hydrogen carbonate by the ammonia process, at Syracuse, New Pork.It was deposited on the inner surface of an iron tank, and had the appearance of boiler-scale, being 1 to 2 inches thick, with a vitreous lustre, and a greenish-grey colour. It is usually covered with crystal planes, proving to be the termination of prisms (probably monoclinic). Analysis gave the following results :- NaC1. Na2C0,. MgCOB. CaC0,. FeCO,. H20. CO,. Total. 22.23 40.62 31.57 3.56 0.08 0.63 0.64 99.33 The scale is evidently a triple sait, represented by the formula MgCO,,Na,?CO,,NaCl. It is, in fact, a definite crystalline product of an interesting constitution, not unlike that of several well -defined mineral species, in which an alkaline chloride appear8 to be in mole- cular combination with heterogeneous materials. By C .HEYER (Ber., 19, 2684-2690).- When strontium hydroxide is heated to bright redness, it is converted into strontium oxide. When the latter is exposed to air saturated with aqueous vapour, and then to dry air at the ordinary femperature, strontia dihydrate, Sr0,2H20, is obtained as a white, crystalline powder. I Dry strontia dihydrate, when heated with dry carhonic anhydride, is completely converted into carbonate ; the monohydrate, however, absorbs only traces of carbonic anhydride. The water in the dihy- drate was determined by passing dry carbonic anhydride over t h e substance for five hours at 26*5", and for 40 minutes at 121.5", and absorbing the water in sulphuric acid bulbs. These results are not in accordance with those obtained by Scheibler (Abstr., 1886, 927).N. H. M. B. H. B. Strontia Dihydrate. Calcium Borate. By B. BLOTJNT (Chem. News, 54,208-209).- The salt obtained by fusing freshly calcined lime with boric anhydride over a Bunsen burner is CaB,O,. I t is possible that t4he salt obtained on a platinum wire loop before the blowpipe, with large excess of boric anhydride, contains a larger proportion of boric acid. D. A. L. Calcium Ammonium Arsenate and Calcium Arsenates. By C. L. BLOXAM (Chem. News, 54, 268-170 ; 193--194).-1n a previous communication (Abstr., 1886, 920), mention is made of a calcium ammonium arsenate. Various observers differ as to the composition of this salt and as to the amount of water it contains. The amount' of water appears to vary somewhat with the state of the atmosphere, and hence, prohably, the cause of the difference of opinion on this point.It is now shown theoretically and experimentally that the* precipitate produced by arsenic acid in a solution of calcium chloride containing free ammonia hae the following composition :-INOROANI C CHEMISTRY. 109 Air-dried.. ........................ CaNHAsSOh + 7H20. Dried in a vacuum over sulphuric acid, Ca3NH4HZ(As04), + 3H20. Dried at 100". ..................... Ca6NH4H6(AsO& + 3H,O. Ignited ........................... CazAs,Oy. I t is suggested to use the precipitation of calcium as calcinm ammo- nium arsenate for quantitative purposes ; it ie convenipnt, the precipi- tate being crystalline and bulky, but is not susceptible of such great accuracy as the oxalate method.It is recommended for checking hardness determinations in water analysis. Its various advantages and disadvantages as a method aye discussed. On evaporating down a hydrochloric acid solution of calcium ammonium arsenate with platinic chloride, the platinochloride after ignition was observed t o be mixed with fine, white, opaque, prismatic crystals of the orthoarsenate Ca3(As04)2, insoluble in acids. A repetition of the experiment re- sulted in the production of a substance somewhat similar in appearance, namely the meta-arsenate C ~ ( A S O ~ ) ~ ; the same substance is formed when mixtures of arsenious anhydride and calcium carbonate are ignited, and is left as an insoluble, crystalline powder when the ignited mass is treated with hydrochloric acid.Artificial Lead Silicate from Bonne Terre, Missouri. By H. A. WHEELER (Amer. J. Sci., 32, 272--273).-E. S. Dana and S. L. Penfield hare given (Abstr., 1886, 317) some crystallographic deter- minations and apalyses of this artificial mineral from the Desloge Lead Co., of Boxine Terre. Since the publication of that paper, the author has examined some specimens in the metallurgical collection of Washington University. The results of his analysis are as follows :- D. A. L. Si02. PbO. Fe203. Al20,. CaO. MgO. I .... 17.11 73.66 0.80 0.53 2.35 0.22 I1 .... 18.51 '72.93 1.31 0.62 1.66 0.201 C1. Nn20. Ni0. Total. I .... 0.08 2.22 3.06 100.03 I1 .... undet. undet. undet. 95.23 I, coarse crystals; 11, fine crystals. The iron in these analyses was assumed to be in the form of ferric oxide.B. H. B. Equivalent of Gadolinium Oxide. By A. E. NORDENSKIOLD (Cowzpt. rend., 103, 795-798).-Gadolinium oxide is the mixture of jttrium, erbium. and ytterbium oxides, which was first obtained from the gadolinite found at Ytterby. It is precipitated by ammonia and ammonium oxalate as well as by potassium sulphate, and the three constituents cannot be separated quantitatively. The gadolinium oxide obtained from kainosite, the silicocarbonate of yttrium, erbium, and ytterbium, recently discovered at Hittero, in Norway, has the molecular weight 260 2 it 0 = 16 and the formula of the oxide is taken as &03. This number is practically identical with the molecular weight of the similar mixture of oxides obtained VOL.LII. i110 ABSTRACTS OF CHEMICAL PAPERS. by different observers (Nordenskiold, Lindstrom, Engstrom, Cl b e ) from gadolinite, kainosite, azzhenitc, xenotime, fergusonite, clevite, fluocerite, and eudialite. These minerals are found in different localities, and contain the oxides in combination with different acids, such as silicic, pbosphoric, niobic, or tantalic acid. Moreover, the oxides have been separated by somewhat different methods, and yet in all cases the greatest variation from the mean value for the mole- cular weight, 261.9, is one per cent., a variation which is within the error of experiment, and is not greatel. than the alterations which have been made in recent times in the atomic weights of some of the better known elements. It follows therefore that gadolinium oxide, although not thc ozide of a simple substance, but a, ?nixtwe of three isomorphous oxides, has a constant molecular weight, even whrrt obtained .from totally &#went minerals found in widely separated localities.This is the first instance of the coexistence of three isomorphous substances in constant DroDortions. The exdanation of this fact seems to be a problem andogius to that of the brigin of the minor planets. C. H. B. Formation of Ultramarine in the Wet Way. By I?. KNAPP ( J . 237'. Chem. [ d ] , 34, 328-340).-An account of some further ex- periments on the formation of ultramarine by the exposure of a heated mixture of kaolin, soda, and sulphur, to a damp atmosphere, or treat- ment of the same with liver of sulphur (Abstr., 1886, 306).The various conditions necessary for success are discussed in full, such as the degree of aggregation of the liver of sulphur and the form of silicate or silica used. Thus experiments with quartz were unsuc- cessful, and those with silicic acid jelly from soluble glass led to the production of it bluish-green material, which turned to a deep blue on warming. Pure alumina led to no result, but sodium aluminate gave a very satisfactory product. Salts of sodium, such as the thio- sulphate, or even calcium phosphate, produced very fine specimens of ultramarine-blue. V. H. V. Sodium Dichromate. By A. STANLEY (Chem. News, 54,194- 196).-Sodium dichromate crystallises wi1,h 2 mols. HzO, in prisms and plates belonging to the triclinic system; its sp. gr.is 2.5246 at 13"; i t is deliquescent. It loses 1 mol. H,O below 75", and all below loo", leaving a light brown, anhydrous salt, which fuses to a transparent dark red liquid a t 320°, and on cooling cr~stallises in the same forms as the hydrated salt, When treated with water, the anhydrous salt causes a rise, and the hydrated salt a fall in temperature. 100 parts of the saturated aqueous solution contain- Temperature.. 0" 15" 30" 80" 100" 139" Parts Na&r2O7 107.2 109.2 116-6 1428 162.8 209.7 The saturated solution boils at 139". A table of the sp. gr. of soh- tious of various strengths is given. Sodium dichromate is insoluble in ether, slightly soluble in alcohol. I t is very hygroscopic, in 48 hours an exposed sample absorbed one-third its weight of water ; a sample of calcinm ch1or:de under fiimilar circumstances absorbedINORGANIC CHEMISTRY.111. nearly its own weight of water. It decomposes slightly above its melting point, and a t a dull red heat leaves sodium chromate and chromic oxide, In its reactions generally it resembles potassium dicliromat e. By dissolving the dichromate in warm aqueous chromium trioxide, the trichromate separates on cooling in dark red crystals ; these a r e very solnble in water, and are deliquescent. By the action of sodium dichromate on freshly prepared cuprio hydroxide, a brown powder is formed, consisting of microscopic crystals of the composition Na2Cr04,CuCrz07,2Cu0 + 4H20- It is almost insoluble in water, and but slightly soluble i n alcohol. It loses all its water below loo", and partially decomposes a t higher tempera; tures, cupric oxide separating.Magnesium sodium chromate, prepared by neutmlising sodium dichromate with magnesia, crystallises with 3 mols. H20, in yellow four-sided prisms and plates. It is soluble in water and alcohol, but insoluble in ether. The 3 mols. H,O are driven off below 200°, leaving a dark reddish-brown powder, which fuses at a red heat with partial decomposition. D. A. L. Reduction of Inorganic Thio-salts by Hydrogen.. By G. K R ~ ~ S S and H. SOLEREDER (Bey., 19, 2729--2759).--By the reduction of molybdothio-salts by hydrogen at a red heat, salts of a lower molybdosulphide could not be obtained ; potassium thiomolybdate and ammonium dithiomolybdate are gradually redaced t o a mixturn of molybdenum and potassium sulphide and to molybdenum.I n the case of ammonium dithiomolybdate, no oxpulphide could be obtained. Potassium thallium sulphide, K2TlaS4, behaves k a similar manner, being reduced to potassium sulphide and thallium. The salts of such thio-acids as remain unchanged in a current of hydrogen are not reduced; for example : R2S,3ZnS ; K,S,SCdS, &3. If the thio-acid is merely reduced to -a lower stable sulphide, the corresponding thio-salt yields also the lower sulphide or its thio-salt in the case of its having the properties of a thio-acid. Examples : K2S,3Cu2S,Cu,S2, K2S,Fe2S3. Atomic Weight of Tungsten. By J. WADDBLL (Amer. Chem. J., 8, 280--288).--ScheeIite was decomposed by nitric acid, and the impure tungsten trioxide freed from silica as follows :-The mixture is fused with an equal weight of hydrogen potassium sulphate until in a state of limpid fusion; the cool mass is digested with water and filtered ; water containing ammonium carbonate must be used for the further washing to keep the fluid from running through turbid, and to assist in the solution of the tnngstate.If the silica is to be esti- mated it will probably require a second fusion, &c., with acid potas- sium sulphate. Another method of separating tungstic acid and silica is by means of ammonia, but here also the extraction is apt $0 be incomplete. Commercial metallic tungsten is but slowly attacked by acids, it cannot be fused with nitre without attacking the crucible, and is therefore best oxidised by a current of air a t ared heat.The product is fused with sodium carbonate, and after dissolution N. H. M. i 2112 ABSTRACTS OF CHEMICAL PAPERS. filtered from uaoxidised material. The filtrate boiled with ammonium carbonate gave a smail precipitate of alumina. To remove molyb- denum, tartaric acid is added (half the weight), then hydrochloric acid (no tungstic acid is precipitated), and sulphuretted hydrogen passed. The blue filtrate is decolorised by a current *of air, and the tungsten separated in 11 fractions by boiling with hydrochloric acid. Fractions 3, 7, and 10 were purified for determination of the atomic weight by solution in ammonia, reprecipitation by hydrochloric acid, and roasting of the precipitate in a current of air. The analysis was effected by heating gradually in a current of hydrogen, using a porcelain boat and tube, and a Fletcher’s blast furnace. No.3 gave W = 184.50, No. 7 184.00, and No. 10 183.67, when 0 = 16. The first number is to be considered most accurate, namely, W = 184.50, 0 = 16; or W = 184.G4, 0 = 15-96. The specific gravity of this sample of metal is 18.77. Compounds of Gold and Nitrogen. By I?. RASCHICI (Annnlen, 235, 341-368).--Aurous oxide, Au20, is most conveniently prepared by boiling a solution of mercurous nitrate with an excess of auric chloride, as descritled by Figuier (Ann. Chim. Phys. [3], 11, 336). The precipitate dried over phosphoric anhydride contains 2 11301s. HzO. When strong ammonia is added to water containing aurous oxide in suspension, a black explosive powder, NAu3,NH3, is obtained. Boiling with water, or with dilute acids, converts the sespui-auroamine into triauramine, NAu3.Met hylamine also acts on aurous oxide, forming diauromethy Zamine, NM eAuz. This compound also loses half its nitrogen when boiled in water for 15 minutes. Gold monoxide, AuO, was first obtained in a pure state by Schottlander (Abstr., 1883, 853). It is converted by strong ammonia into the very explosive compound sespu ihydrazhry Zamine, NH,,N( AuOH)~. Warm hydro- chloric acid dissolves two-thirds of the gold, but one-third is left in the metallic state. On boiling with water, half the nitrogen is expelled, and trihydraurylamine remains. Methylamine unites with gold monoxide, formirig the compound MeN(AuOH)a ; this loses half its nitrogen in the form of methylamine when boiled in water. The author confirmsthe accuracFof Dumas’s formula, (AuN + NH3)2+ 3H20, for the fiilminating gold prepared by the action of ammonia on auric oxide, Auz03 ; but he proves t h a t t<he substance which is precipitated by ammonia from a solution of auric chloride is a mixture of the preceding compound and auric imidochloride, NH : AuCl.The latter compound is obtained in an impure state when a, hew drops of ammonia are added to a large excess of auric chloride. Only a por- tion of the chlorine can be removed by digestion with ammonia. Fulminating gold is slowly attacked by boiling water, losing ammonia. Boiling with nitric acid renders the compound more explosive. Dilute sulphuric acid does not liberate ammonia, but forms a very explosive substance of the composition ( AuN2H3),,H28O4.The constitution of auric diamine is represented by the formula NH,*Au : NH. The decomposition it undergoes when boiled with water is probably 2NHzAuNH = NH : AuN Methylamine yields a, yellow precipitate with auric chloride, soluble H. B. AuNH2 -I- NHs.INORGAN10 UHEMISTRT. 113 in an exoess of the reagent. On warming the solution, a bibown pre- cipitate which contains a large quantity of chlorine is deposited. Auric oxide yields a red compound with methylamine. Complex Inorganic Acids. By W. GIBBS (Amer. Chem. J., 8, 289-290 j .-Compounds have been obtained comparable with the chlorplatinophosphoric acids of Schutzenberger. These compounds are to be regarded as phosphoric acid in which one atom of oxygen is replaced by platinous chloride, bromide or oxide, or the corresponding compounds of palladium, iridium, ruthenium, and osmium.More- over the phosphorus may be replaced by arsenic or antimony. The following formula? are given to the compounds so far studied :- w. c. w. As,0,3( As203,2Pt0),5 (NH4)aO + 7H20 ; fAs,O3,2PtCI,)K,O + 2HzO ; ~ ~ W O ~ , A S ~ ~ ~ , ~ ( A S ~ O ~ , ~ P ~ C ~ , ) , ~ ~ P ~ ( + 60R2O j 23 W O ~ , ~ A S ~ O ~ , ~ ( A S ~ O ~ , ' ~ P ~ O ) 6Pt (N&),O + 4OH2O ; ~~XOO~,AS~O~(AS~O~,~P~O)~P~(NH~)~O + 27H20. H. B. Roseo-rhodium Salts. By S. M. JORGENSEN (J. pr. Chem. [2], 34, 394-406 ; compare Abstr., 1883, 1058) .-Boseo-rhodium nitrate, ( Rh2,10XH,,2Hz0) (NO,), is obtained by adding moderately dilute nitric acid to a concentrated solution of roseo-rhodium hydroxide, when i t separates as a white precipitate consisting of microscopic, quadratic tables.It loses 2 mols. of water at loo", and is converted into nitratopurpureo-rhodium nitrate. Roseo-rhodiunz platinochloride witrate, ( Rh2,1 0NH3,2H,0)(N0,),PtC16 + 2H20, is obtained as a beautiful, orange-yellow, crystalline pre- cipitate on adding hydrogen platinochloride to a solution of the nitrate. It emily loses 2 mo1s. of water at 100°, the other 2 mols. are given off more slowly, and nitratopurpureo-rhodium platinochloride is formed; on treating this with dilute hydrochloric acid all the platinum is dissolved, leaving white nitratopurpurco-rhodium chloride undissolved. A solution of the latter in water give8 wit4 sodium dithionate a precipitate of the characteristic nitratopurpureo-dithionate.Roseo-rhodium bromide, ( Rh,,10NH3,2H20j,Br6, is prepared by adding concentrated hydrogen bromide to a solution of roseo-rhodium hydroxide. It forms a crystalline powder consisting of small octo- hedrons or six-sided tables. At loo", it loses 2 mols. H20 a.nd is converted into the bromopurpureo-bromide. It is easily soluble i n cold water, and its aqueous solution gives the same reactions as Che nitrate. Roseo-rhodium sutphate, ( Rhz,10NH3,2E20) (SO& + 2H20, is pre- pared by saturating the hydroxide with dilute sulphuric acid, and precipitating with alcohol. It forms a white precipitate of very small octohedral crystals. 'In contrast to the other roseo-rhodium salts, i t is not converted into the purpureo-salt by solution in boiling water ; i t crystallises in large, quadratic prisms, terminated by a quadratic pyramid, apparently isomorphous with roseocobalt sulphate.At looo, i t quickly loses 4 mols. H20, and then more slowly another 4 mol., undergoing a similar ahanga to the analogous roseooobitlt salt, form- It is easily soluble in water.114 ABSTRACTS OF CHEMICAL PAPERS. ing luteo-salts. Bn aqueous solution gives all the reactions of the nitrate ; with potassiiim iodide, it gives a white precipitate, consisting of microscopic octohedrons of the roseoiodide sulphate, (Rh,,lONH,,2HzO)I,(SO,)2. Roseo-rhod ium platinochloride sdyhnte, ( Rhz, 1 O?SH3,2H20)PtC16( SO,)?, is obtained as ti beautiful, silky, buff precipitate of very thin six-sided tables, by adding a solution of hydrogen platinochloride to a cold solution of the roseo-sulphate. It seems to be isomorphous with the corresponding roseo- and luteo-salts of cobalt and chromium.Roseo-rhodium orthophoyhzte, ( Rhz,10NH3,2H20) (PO4H)s + 4Hzo, is prepared by adding a 10 per cent. solution of phosphoric acid to a concentrated solution of the hydroxide. It forms a white, crystalline powder which behaves like the correFponding cobalt salt. Sodium roseo-rhodium pyrophosphate, ( Rhz,10NH3,2H,0),P20,Na + 23H20, is obtained by adding sodium pyrophosphate to a solution of the hydroxide iu dilute hydrochloric acid until the amorphous precipitate a t first formed is redissolved ; on shaking, the salt then separates as a silky, white crystalline precipitate. It agrees in all respects with the cobalt salt.Koseo-rhodium cobalticyanide, (Rh,,10NH3,2HzO),COzCyn, is obtained a s srnall pale-yellow crystals on adding a solution of potassium cobalti- cyanide to the hydroxide, neutralised with dilute sulphuric acid. Nitratopurpureo-rhodium Salts. By S . M. JORGEXSEN ( J . pr. G. H. M. C'hem. ['L], 34, 40 -109) .-Nitratopur~ureo-rhodiunz nitrate, is easily obtained by heating the roseonitrate or by dissolving this salt in hot water, adding an equal volume of Concentrated nitric acid, heating the mixture for some time on the water-bath and then allow- ing it to cool, when the salt separates in small, octohedral, white crystals with a greenish-yellow tinge. Heated over a flame, the salt explodes. Nitrat~urpureo-rhoa~u~n chloride, (K 0,) L( Rhp, 10NH,)Cl,, is obtained as a pale greenish-yellow precipitate on filtering a, cold solution of the nitrate into an ice-cold mixture of 3 vols.hydrochloric acid and 1 vol. water. Nitratopurpureo-rhodium dithionate, (NO,),( Rhz,10NH3)( Sz06)2 + 2H20, is prepared by precipitating a cold saturated solution of the nitrate or chloride with sodium dithionate. It forms white, silky X-shaped aggregations resembling the corresponding cobalt salt. The air-dried salt loses all its water a t 100". It is quite insoluble i n water. G. H. M. It is only slightly soluble in water. An aqueous solution gives all the reactions of the nitrate. Xantho-rhodium Salts. By S. M. J~SKGENSEN (J. pr. Chem. [2], 34, 410-423) .-Xantho-rhodium nitrate, (NOz),( Rhs, 10NH)(3N03)4, is pre- pared by dissolving the chloropnrpureochloride in dilute sodium hydroxide on the water-bath, and then adding pure sodium nitrite and dilute nitric acid ; the salt separates as a, white, crystalline powder which can be recrystallised from hot water.Heated over a flame, theINORWNIC CHEMISTRY. 115 salt explodes and leaves a voluminous residue of rhodium. The xanthonitrate is fairly soluble in cold, easily in hot water ; insoluble in alcohol. Its aqueous solution, unlike that of the nitratonitrate, is not converted into the roseonitrato by boiling. It is also unacted on by sodium hydroxide, ammonium sulphide, &c. Xantho-rhodium chloride. (NO,),(Rh,,lUNH,) C14, is prepared in the same way as the nitrate, adding dilute hydrochloric acid, however, instead of nitric acid after the sodium nitrite, when the salt separates RS a white powder ; or from the nitrate, by filtering a concentrated solution of the latter into dilute hydrochloric acid, the addition of alcohol then throws down a yellowiBh-white precipitate of small octa- hedrons.When heated, the salt is decomposed, and leaves a residue of pure rhodium. Treated with silver oxide and water, the chloride yields a solution oExantho-rhodium hydroxide, which behaves as a strong a1 kali. Xantho-rhodium platinochloride, (NO&( Rhz,10NH3) (PtC16)2, is ob- tained as a buff -colonred precipitate consisting of microscopic needles by adding hydrogen platinochloride to a solution of the nitrate. This salt is analogous to the corresponding cobalt and chromium salts. Xuntho-rhodium bromide, (NOz),( Rhz,10NH3)Rr4, is obtained as a white, crystalline preciphate on adding concentrated hydrobromic acid to a cold solution of the nitrate.I t is easily soluble in water, and the solution gives all the reactions of the nitrate. Xardho-rhodium dithionnte, (NO,),( Rh,,lONH,)(S,O,), + 2H20, is prepared by filtering a cold solution of the nitrate into an excess of sodium dithionate. It forms a white, crystalline precipitate consisting of microscopic prisms. Xuntho-rhodium sulphutes are prepared by treating the chloride with strong sulphuric acid, &c. (a) The normal salt, It loses all its water at 100". (NOz) 2 ( Rh2,10NH3) ( S O&, crystallises in Bat, brilliant, needles many centimetres long. It may also be obtained by precipitating a solution of the acid salt with alcohol.The salt explodes on heating. With a solution of iodine in potassium iodide, it yields a periodide. ( b ) The acid salt, 2[ (NO,),(€th,,lONH,)( SO&],( H2S04),, crystallises in long, white needles. The salt decomposes quietly on heating, leaving the rhodium in pseudomorphs of the crystals. Xuntho-rhodium siZicoJZuoride, (NO,),( Rh,,lONH,)( SiF6),, is obtained AS a snow-white, silky precipitate on adding a cold solution of the nitrate to an excess of hydrofluosilicic acid. Under the microscope, i t consists of small rhombic tables resembling the chloropurpureo- rhodium silicofluoride. Dilute nitric acid decomposes it, forming the nitrate and free acid. Xatztho-rhodium oxaZate, (NO,),( Rh2,10NH3) ( C,O,),, is obtained as a white precipitate consisting of microscopic prisms, by adding a solu- tion of the nitrate to excess oE ammonium oxitlate.Nitric acid decomposes it in the same way as the silicofluoride. The roseo- and purpureo- as well as the xantho-salts of rhodium show a most unmistakable resemblance to those of cobalt and chro- miurn. The xantho-rhodium salts are, however, much mom stable116 ABSTRACTS OF CHEMICAL PAPERS. than those of cobalt and chromium. The nitrate is especially so, showing that the xantho-salts are really nitritopurpureo-salts. The great stability of the nitrate is partly accounted for by the nature of the metal, and also probably partly by the fact that pentad nitrogen is combined with pentad nitrogen, thus :- 0 : N.0.106 ABST tiACTS OF CHEMICAL PAPERS.In o r g a n i c C h e m i s t r y.Action of Hypochlorous Anhydride on Iodine Trichloride.Ry H.BASSETT and E. FIELDING (Chem. News, 54, 205--206).--Iodicanhydride is the main product of the reaction when solutions ofiodine trichloride and’ hypochlorous anhydride in carbon tetrachlorideare mixed, or when a current of bypochlorous anhydride is passedthrough a solution of iodine trichloride, or even over solid trichloride.In the last case, the action is very slow. D. A. L.Saturation of Selenious Acid by Bases. By C. BLAREZ (Conzpt.rend., 103, 804- 806). -With cochineal or helianthin (methyl-orange), selenious acid is monobasic. With litmus, it is monobasic toammonia, lime, strontia, and baryta, but if soda or potash is used thelitmus only becomes blue-violet when about 1.5 equivalent of alkaliis added.With phendphthalein. and potash, soda, ammonia, lime orstrontia, the colour change takes place when somexhnt more than1.5 equivalent of alkali is added, but with phenolphthalein and barytaselenious acid is bihasic,, and the colour change is sharp and distinct.Selenious acid can be accurately titrated by means of standardbarytn solution, using helianthin or phenolphthalein as indicator.The bavyta has double the value with the first indicator that i t haswith the second. Both indicators can be used in the same solution ;the rose colour of the heliauthin disappears when the acid is halfsaturated, and the rose colour due to the change of the phenol-phthalein becomes visible when saturation ik complete.As no.basic barium selenite is formed when baryta is present inexcess, an excess of baryta can be determined by means of phenol-phthaleh and a standard acid in the liquid containing the precipitatedbarium selenite.Selenious acid, like sulphurms acid, cant be estimated in presenceof other acids by means of baryba, provided that the tofal basicity ofthese acids is indicated by helianthin. Helianthin and phenol-phthaleln are both a;dded to the same solution, and the amount ofselenious acid is calculated from the quantity of bnryta solutionrequired to produce the second colour change, this quantity beingexactly half that which would be required to neutralise the seleniousacid alone with phenolphthale’in as the indieator. C. H, B.Formation of Nitrites- By S.KAPPEL (Arch. Pkarm. [3], 24,897--900).--The author has extended his observations made on theaction of copper, iron, and zinc in contact with the air, and solutionsof ammonia and the fixed alkalis (AbsBr., ’1883, 282, 286), employingin recent experiments magnesium, aluminium, and tin. Magnesiumexposed to the a i r in contact with aqueous potash, gave small quan-tities of nitrous acid, ozone, and hydrogen peroxide. With ammoniasolution, strong indications of nitrous acid were obtained. Alnmiriiumin potassium hydrate produced nitrous acid readily, even in the coldINORGAKIC CHEMISTRY. 107nitric acid and hydrogen peroxide were also easily detected. Withammonia the action was much slower. Tinfoil gave scarcely anyreaction ; nitrous acid was not produced, but hydrogen peroxide wasperceptible. J.T.Compounds of Arsenious Anhydride with HaTogen Salts.By F. R~UORFF (Ber., 19,2668--2679).-1n this paper the preparationand properties of compounds of the halogen salts of the alkali metalswith arsenious anhydride are described. These are best obtained bypassing carbonic anhydride info a mixed solution of potassium arseniteand the halogen salt. These componnds separate either in the amor-phous form or in indistinct crystals, of the general formula MX,2As203,sparingly soluble in water, insoluble in alkaline carbonates, but verysoluble in the free alkalis; when heated, they decompose with elimi-nation of arsenious anhydride, the compound KI, 2As,03 decomposinga t 350" ; KBr,2As20z a t 380" ; and KCI,As203 a t 240" ; a compound,KC1,2Asz0,, is also described.The ammonium compoundsNH41,AsL03, NH4Br,2As,C3, and NH4Cl,As20,, are also described,those of sodium being reserved for a future communication.V. H. V.Some Probable New Elements. By A. PRINGLE (Chem. News,54, 167-168).-TThe author claims to have discovered some new sub-stances, including five metals and a substance resembling selenium,called hesperisiztm, in some " Lower Silurian " rocks, situated in thecounty of Selkirk. One metal is said to be like iron, but gives neitherthe thiocyanate nor the tannic acid reaction ; one is Eke lead in appear-ance, is easily fused and volatilised, and yields yellow and green salts ;another which is charcoal-black, is called erebodium,; its equivalent is95.4, and it forms several oxides. A fourth, g a d e n i m , with equivalentabout 43.6, a light-grey powder, forms a red monoxide and a cream-coloured dioxide, yielding respectively white and yellow saltp.Another, polymnesturn (Pm), is a rather dark-coloured metal, withequivalent about 74.A preliminary description of four oxides, PmO,PmOs, PmO,, and (?)I Pm05, of two sulphides, PmS and PmSp, andof other compounds is given. D. A. L.Production of Alkali Metals. By H. Y. CASTXER (Chem. News,54, 218--SlS).--Irou reduced by hydrogen or carbonic oxide ismixed with t a r in proper proportions, so that after the mixture iscoked it has a composition about = FeCp The coke is finely ground,mixed with c a s t i c soda (or potash) in proportions = 3NaHO +E'eC,, or about 100 NaHO to 15 of coke. This mixture is introducedinto a cast-iron crucible, and heated in a specially arranged furnace(described in the paper).The reduction and distillation commence ata temperatnre of 1000". When the operation is finished the crucibleis removed, and another immediately put in its place. The residueconsists of some sodium carbonate, and finely divided iron; thesodium carbonate is recovered, the iron used to make fresh reducingcoke, aiid the crucible is used over and over again. The advantagesover the old method are manifest. D. A. L108 ABSTRACTS OF CHEMICAL PAPERS.Crystalline Scale formed in the Manufacture of SodiumHydrogen Carbonate. By G. W. LKIGHTON (Amer. .J.Xci., 33,318--.319).-This scale was formed in the manufacture of sodiumhydrogen carbonate by the ammonia process, at Syracuse, New Pork.It was deposited on the inner surface of an iron tank, and had theappearance of boiler-scale, being 1 to 2 inches thick, with a vitreouslustre, and a greenish-grey colour. It is usually covered with crystalplanes, proving to be the termination of prisms (probably monoclinic).Analysis gave the following results :-NaC1. Na2C0,. MgCOB. CaC0,. FeCO,. H20. CO,. Total.22.23 40.62 31.57 3.56 0.08 0.63 0.64 99.33The scale is evidently a triple sait, represented by the formulaMgCO,,Na,?CO,,NaCl. It is, in fact, a definite crystalline product ofan interesting constitution, not unlike that of several well -definedmineral species, in which an alkaline chloride appear8 to be in mole-cular combination with heterogeneous materials.By C .HEYER (Ber., 19, 2684-2690).-When strontium hydroxide is heated to bright redness, it is convertedinto strontium oxide. When the latter is exposed to air saturatedwith aqueous vapour, and then to dry air at the ordinary femperature,strontia dihydrate, Sr0,2H20, is obtained as a white, crystallinepowder.I Dry strontia dihydrate, when heated with dry carhonic anhydride,is completely converted into carbonate ; the monohydrate, however,absorbs only traces of carbonic anhydride. The water in the dihy-drate was determined by passing dry carbonic anhydride over t h esubstance for five hours at 26*5", and for 40 minutes at 121.5", andabsorbing the water in sulphuric acid bulbs.These results are notin accordance with those obtained by Scheibler (Abstr., 1886, 927).N. H. M.B. H. B.Strontia Dihydrate.Calcium Borate. By B. BLOTJNT (Chem. News, 54,208-209).-The salt obtained by fusing freshly calcined lime with boric anhydrideover a Bunsen burner is CaB,O,. I t is possible that t4he salt obtainedon a platinum wire loop before the blowpipe, with large excess ofboric anhydride, contains a larger proportion of boric acid.D. A. L.Calcium Ammonium Arsenate and Calcium Arsenates. ByC. L. BLOXAM (Chem. News, 54, 268-170 ; 193--194).-1n a previouscommunication (Abstr., 1886, 920), mention is made of a calciumammonium arsenate. Various observers differ as to the compositionof this salt and as to the amount of water it contains.The amount'of water appears to vary somewhat with the state of the atmosphere,and hence, prohably, the cause of the difference of opinion on thispoint. It is now shown theoretically and experimentally that the*precipitate produced by arsenic acid in a solution of calcium chloridecontaining free ammonia hae the following composition :INOROANI C CHEMISTRY. 109Air-dried.. ........................ CaNHAsSOh + 7H20.Dried in a vacuum over sulphuric acid, Ca3NH4HZ(As04), + 3H20.Dried at 100". ..................... Ca6NH4H6(AsO& + 3H,O.Ignited ........................... CazAs,Oy.I t is suggested to use the precipitation of calcium as calcinm ammo-nium arsenate for quantitative purposes ; it ie convenipnt, the precipi-tate being crystalline and bulky, but is not susceptible of such greataccuracy as the oxalate method.It is recommended for checkinghardness determinations in water analysis. Its various advantagesand disadvantages as a method aye discussed. On evaporating downa hydrochloric acid solution of calcium ammonium arsenate withplatinic chloride, the platinochloride after ignition was observed t o bemixed with fine, white, opaque, prismatic crystals of the orthoarsenateCa3(As04)2, insoluble in acids. A repetition of the experiment re-sulted in the production of a substance somewhat similar in appearance,namely the meta-arsenate C ~ ( A S O ~ ) ~ ; the same substance is formedwhen mixtures of arsenious anhydride and calcium carbonate areignited, and is left as an insoluble, crystalline powder when the ignitedmass is treated with hydrochloric acid.Artificial Lead Silicate from Bonne Terre, Missouri.By H.A. WHEELER (Amer. J. Sci., 32, 272--273).-E. S. Dana and S. L.Penfield hare given (Abstr., 1886, 317) some crystallographic deter-minations and apalyses of this artificial mineral from the DeslogeLead Co., of Boxine Terre. Since the publication of that paper, theauthor has examined some specimens in the metallurgical collectionof Washington University. The results of his analysis are asfollows :-D. A. L.Si02. PbO. Fe203. Al20,. CaO. MgO.I .... 17.11 73.66 0.80 0.53 2.35 0.22I1 .... 18.51 '72.93 1.31 0.62 1.66 0.201C1. Nn20. Ni0. Total.I ....0.08 2.22 3.06 100.03I1 .... undet. undet. undet. 95.23I, coarse crystals; 11, fine crystals. The iron in these analyseswas assumed to be in the form of ferric oxide. B. H. B.Equivalent of Gadolinium Oxide. By A. E. NORDENSKIOLD(Cowzpt. rend., 103, 795-798).-Gadolinium oxide is the mixture ofjttrium, erbium. and ytterbium oxides, which was first obtainedfrom the gadolinite found at Ytterby. It is precipitated by ammoniaand ammonium oxalate as well as by potassium sulphate, and thethree constituents cannot be separated quantitatively.The gadolinium oxide obtained from kainosite, the silicocarbonateof yttrium, erbium, and ytterbium, recently discovered at Hittero, inNorway, has the molecular weight 260 2 it 0 = 16 and the formulaof the oxide is taken as &03.This number is practically identicalwith the molecular weight of the similar mixture of oxides obtainedVOL. LII. 110 ABSTRACTS OF CHEMICAL PAPERS.by different observers (Nordenskiold, Lindstrom, Engstrom, Cl b e )from gadolinite, kainosite, azzhenitc, xenotime, fergusonite, clevite,fluocerite, and eudialite. These minerals are found in differentlocalities, and contain the oxides in combination with different acids,such as silicic, pbosphoric, niobic, or tantalic acid. Moreover, theoxides have been separated by somewhat different methods, and yetin all cases the greatest variation from the mean value for the mole-cular weight, 261.9, is one per cent., a variation which is within theerror of experiment, and is not greatel.than the alterations whichhave been made in recent times in the atomic weights of some of thebetter known elements. It follows therefore that gadolinium oxide,although not thc ozide of a simple substance, but a, ?nixtwe of threeisomorphous oxides, has a constant molecular weight, even whrrt obtained.from totally &#went minerals found in widely separated localities. Thisis the first instance of the coexistence of three isomorphous substancesin constant DroDortions. The exdanation of this fact seems to be aproblem andogius to that of the brigin of the minor planets.C. H. B.Formation of Ultramarine in the Wet Way. By I?. KNAPP( J . 237'. Chem. [ d ] , 34, 328-340).-An account of some further ex-periments on the formation of ultramarine by the exposure of a heatedmixture of kaolin, soda, and sulphur, to a damp atmosphere, or treat-ment of the same with liver of sulphur (Abstr., 1886, 306).Thevarious conditions necessary for success are discussed in full, such asthe degree of aggregation of the liver of sulphur and the form ofsilicate or silica used. Thus experiments with quartz were unsuc-cessful, and those with silicic acid jelly from soluble glass led to theproduction of it bluish-green material, which turned to a deep blueon warming. Pure alumina led to no result, but sodium aluminategave a very satisfactory product. Salts of sodium, such as the thio-sulphate, or even calcium phosphate, produced very fine specimens ofultramarine-blue. V. H. V.Sodium Dichromate. By A.STANLEY (Chem. News, 54,194-196).-Sodium dichromate crystallises wi1,h 2 mols. HzO, in prisms andplates belonging to the triclinic system; its sp. gr. is 2.5246 at 13"; i tis deliquescent. It loses 1 mol. H,O below 75", and all below loo",leaving a light brown, anhydrous salt, which fuses to a transparentdark red liquid a t 320°, and on cooling cr~stallises in the same formsas the hydrated salt, When treated with water, the anhydrous saltcauses a rise, and the hydrated salt a fall in temperature. 100 partsof the saturated aqueous solution contain-Temperature.. 0" 15" 30" 80" 100" 139"Parts Na&r2O7 107.2 109.2 116-6 1428 162.8 209.7The saturated solution boils at 139". A table of the sp. gr. of soh-tious of various strengths is given.Sodium dichromate is insolublein ether, slightly soluble in alcohol. I t is very hygroscopic, in48 hours an exposed sample absorbed one-third its weight of water ;a sample of calcinm ch1or:de under fiimilar circumstances absorbeINORGANIC CHEMISTRY. 111.nearly its own weight of water. It decomposes slightly above itsmelting point, and a t a dull red heat leaves sodium chromate andchromic oxide, In its reactions generally it resembles potassiumdicliromat e.By dissolving the dichromate in warm aqueous chromium trioxide,the trichromate separates on cooling in dark red crystals ; these a r every solnble in water, and are deliquescent.By the action of sodium dichromate on freshly prepared cupriohydroxide, a brown powder is formed, consisting of microscopiccrystals of the composition Na2Cr04,CuCrz07,2Cu0 + 4H20- It isalmost insoluble in water, and but slightly soluble i n alcohol.It losesall its water below loo", and partially decomposes a t higher tempera;tures, cupric oxide separating.Magnesium sodium chromate, prepared by neutmlising sodiumdichromate with magnesia, crystallises with 3 mols. H20, in yellowfour-sided prisms and plates. It is soluble in water and alcohol, butinsoluble in ether. The 3 mols. H,O are driven off below 200°,leaving a dark reddish-brown powder, which fuses at a red heat withpartial decomposition. D. A. L.Reduction of Inorganic Thio-salts by Hydrogen.. By G.K R ~ ~ S S and H. SOLEREDER (Bey., 19, 2729--2759).--By the reductionof molybdothio-salts by hydrogen at a red heat, salts of a lowermolybdosulphide could not be obtained ; potassium thiomolybdate andammonium dithiomolybdate are gradually redaced t o a mixturn ofmolybdenum and potassium sulphide and to molybdenum.I n the caseof ammonium dithiomolybdate, no oxpulphide could be obtained.Potassium thallium sulphide, K2TlaS4, behaves k a similar manner,being reduced to potassium sulphide and thallium.The salts of such thio-acids as remain unchanged in a currentof hydrogen are not reduced; for example : R2S,3ZnS ; K,S,SCdS,&3. If the thio-acid is merely reduced to -a lower stable sulphide,the corresponding thio-salt yields also the lower sulphide or itsthio-salt in the case of its having the properties of a thio-acid.Examples : K2S,3Cu2S,Cu,S2, K2S,Fe2S3.Atomic Weight of Tungsten.By J. WADDBLL (Amer. Chem. J.,8, 280--288).--ScheeIite was decomposed by nitric acid, and theimpure tungsten trioxide freed from silica as follows :-The mixture isfused with an equal weight of hydrogen potassium sulphate until ina state of limpid fusion; the cool mass is digested with water andfiltered ; water containing ammonium carbonate must be used for thefurther washing to keep the fluid from running through turbid, andto assist in the solution of the tnngstate. If the silica is to be esti-mated it will probably require a second fusion, &c., with acid potas-sium sulphate. Another method of separating tungstic acid andsilica is by means of ammonia, but here also the extraction is apt $0be incomplete.Commercial metallic tungsten is but slowly attackedby acids, it cannot be fused with nitre without attacking thecrucible, and is therefore best oxidised by a current of air a t ared heat.The product is fused with sodium carbonate, and after dissolutionN. H. M.i 112 ABSTRACTS OF CHEMICAL PAPERS.filtered from uaoxidised material. The filtrate boiled with ammoniumcarbonate gave a smail precipitate of alumina. To remove molyb-denum, tartaric acid is added (half the weight), then hydrochloricacid (no tungstic acid is precipitated), and sulphuretted hydrogenpassed. The blue filtrate is decolorised by a current *of air, andthe tungsten separated in 11 fractions by boiling with hydrochloricacid. Fractions 3, 7, and 10 were purified for determination of theatomic weight by solution in ammonia, reprecipitation by hydrochloricacid, and roasting of the precipitate in a current of air.The analysiswas effected by heating gradually in a current of hydrogen, using aporcelain boat and tube, and a Fletcher’s blast furnace. No. 3 gaveW = 184.50, No. 7 184.00, and No. 10 183.67, when 0 = 16. Thefirst number is to be considered most accurate, namely, W = 184.50,0 = 16; or W = 184.G4, 0 = 15-96. The specific gravity of thissample of metal is 18.77.Compounds of Gold and Nitrogen. By I?. RASCHICI (Annnlen,235, 341-368).--Aurous oxide, Au20, is most conveniently preparedby boiling a solution of mercurous nitrate with an excess of auricchloride, as descritled by Figuier (Ann.Chim. Phys. [3], 11, 336).The precipitate dried over phosphoric anhydride contains 2 11301s.HzO. When strong ammonia is added to water containing aurousoxide in suspension, a black explosive powder, NAu3,NH3, is obtained.Boiling with water, or with dilute acids, converts the sespui-auroamineinto triauramine, NAu3. Met hylamine also acts on aurous oxide,forming diauromethy Zamine, NM eAuz. This compound also loses halfits nitrogen when boiled in water for 15 minutes. Gold monoxide,AuO, was first obtained in a pure state by Schottlander (Abstr., 1883,853). It is converted by strong ammonia into the very explosivecompound sespu ihydrazhry Zamine, NH,,N( AuOH)~. Warm hydro-chloric acid dissolves two-thirds of the gold, but one-third is left inthe metallic state.On boiling with water, half the nitrogen isexpelled, and trihydraurylamine remains. Methylamine unites withgold monoxide, formirig the compound MeN(AuOH)a ; this loses halfits nitrogen in the form of methylamine when boiled in water. Theauthor confirmsthe accuracFof Dumas’s formula, (AuN + NH3)2+ 3H20,for the fiilminating gold prepared by the action of ammonia on auricoxide, Auz03 ; but he proves t h a t t<he substance which is precipitatedby ammonia from a solution of auric chloride is a mixture of thepreceding compound and auric imidochloride, NH : AuCl. The lattercompound is obtained in an impure state when a, hew drops ofammonia are added to a large excess of auric chloride. Only a por-tion of the chlorine can be removed by digestion with ammonia.Fulminating gold is slowly attacked by boiling water, losingammonia.Boiling with nitric acid renders the compound moreexplosive. Dilute sulphuric acid does not liberate ammonia, butforms a very explosive substance of the composition ( AuN2H3),,H28O4.The constitution of auric diamine is represented by the formulaNH,*Au : NH. The decomposition it undergoes when boiled withwater is probably 2NHzAuNH = NH : AuNMethylamine yields a, yellow precipitate with auric chloride, solubleH. B.AuNH2 -I- NHsINORGAN10 UHEMISTRT. 113in an exoess of the reagent. On warming the solution, a bibown pre-cipitate which contains a large quantity of chlorine is deposited.Auric oxide yields a red compound with methylamine.Complex Inorganic Acids.By W. GIBBS (Amer. Chem. J., 8,289-290 j .-Compounds have been obtained comparable with thechlorplatinophosphoric acids of Schutzenberger. These compoundsare to be regarded as phosphoric acid in which one atom of oxygen isreplaced by platinous chloride, bromide or oxide, or the correspondingcompounds of palladium, iridium, ruthenium, and osmium. More-over the phosphorus may be replaced by arsenic or antimony. Thefollowing formula? are given to the compounds so far studied :-w. c. w.As,0,3( As203,2Pt0),5 (NH4)aO + 7H20 ;fAs,O3,2PtCI,)K,O + 2HzO ;~ ~ W O ~ , A S ~ ~ ~ , ~ ( A S ~ O ~ , ~ P ~ C ~ , ) , ~ ~ P ~ ( + 60R2O j23 W O ~ , ~ A S ~ O ~ , ~ ( A S ~ O ~ , ' ~ P ~ O ) 6Pt (N&),O + 4OH2O ;~~XOO~,AS~O~(AS~O~,~P~O)~P~(NH~)~O + 27H20.H.B.Roseo-rhodium Salts. By S. M. JORGENSEN (J. pr. Chem. [2],34, 394-406 ; compare Abstr., 1883, 1058) .-Boseo-rhodium nitrate,( Rh2,10XH,,2Hz0) (NO,), is obtained by adding moderately dilutenitric acid to a concentrated solution of roseo-rhodium hydroxide,when i t separates as a white precipitate consisting of microscopic,quadratic tables. It loses 2 mols. of water at loo", and is convertedinto nitratopurpureo-rhodium nitrate.Roseo-rhodiunz platinochloride witrate, ( Rh2,1 0NH3,2H,0)(N0,),PtC16 + 2H20, is obtained as a beautiful, orange-yellow, crystalline pre-cipitate on adding hydrogen platinochloride to a solution of thenitrate. It emily loses 2 mo1s. of water at 100°, the other 2 mols.are given off more slowly, and nitratopurpureo-rhodium platinochlorideis formed; on treating this with dilute hydrochloric acid all theplatinum is dissolved, leaving white nitratopurpurco-rhodium chlorideundissolved.A solution of the latter in water give8 wit4 sodiumdithionate a precipitate of the characteristic nitratopurpureo-dithionate.Roseo-rhodium bromide, ( Rh,,10NH3,2H20j,Br6, is prepared by addingconcentrated hydrogen bromide to a solution of roseo-rhodiumhydroxide. It forms a crystalline powder consisting of small octo-hedrons or six-sided tables. At loo", it loses 2 mols. H20 a.nd isconverted into the bromopurpureo-bromide. It is easily soluble i ncold water, and its aqueous solution gives the same reactions as Chenitrate.Roseo-rhodium sutphate, ( Rhz,10NH3,2E20) (SO& + 2H20, is pre-pared by saturating the hydroxide with dilute sulphuric acid, andprecipitating with alcohol.It forms a white precipitate of very smalloctohedral crystals. 'In contrast to the other roseo-rhodium salts, i t isnot converted into the purpureo-salt by solution in boiling water ; i tcrystallises in large, quadratic prisms, terminated by a quadraticpyramid, apparently isomorphous with roseocobalt sulphate. At looo,i t quickly loses 4 mols. H20, and then more slowly another 4 mol.,undergoing a similar ahanga to the analogous roseooobitlt salt, form-It is easily soluble in water114 ABSTRACTS OF CHEMICAL PAPERS.ing luteo-salts. Bn aqueous solution gives all the reactions of thenitrate ; with potassiiim iodide, it gives a white precipitate, consistingof microscopic octohedrons of the roseoiodide sulphate,(Rh,,lONH,,2HzO)I,(SO,)2.Roseo-rhod ium platinochloride sdyhnte, ( Rhz, 1 O?SH3,2H20)PtC16( SO,)?,is obtained as ti beautiful, silky, buff precipitate of very thin six-sidedtables, by adding a solution of hydrogen platinochloride to a coldsolution of the roseo-sulphate.It seems to be isomorphous with thecorresponding roseo- and luteo-salts of cobalt and chromium.Roseo-rhodium orthophoyhzte, ( Rhz,10NH3,2H20) (PO4H)s + 4Hzo,is prepared by adding a 10 per cent. solution of phosphoric acid to aconcentrated solution of the hydroxide. It forms a white, crystallinepowder which behaves like the correFponding cobalt salt.Sodium roseo-rhodium pyrophosphate, ( Rhz,10NH3,2H,0),P20,Na +23H20, is obtained by adding sodium pyrophosphate to a solution of thehydroxide iu dilute hydrochloric acid until the amorphous precipitatea t first formed is redissolved ; on shaking, the salt then separates as asilky, white crystalline precipitate.It agrees in all respects with thecobalt salt.Koseo-rhodium cobalticyanide, (Rh,,10NH3,2HzO),COzCyn, is obtaineda s srnall pale-yellow crystals on adding a solution of potassium cobalti-cyanide to the hydroxide, neutralised with dilute sulphuric acid.Nitratopurpureo-rhodium Salts. By S . M. JORGEXSEN ( J . pr.G. H. M.C'hem. ['L], 34, 40 -109) .-Nitratopur~ureo-rhodiunz nitrate,is easily obtained by heating the roseonitrate or by dissolving this saltin hot water, adding an equal volume of Concentrated nitric acid,heating the mixture for some time on the water-bath and then allow-ing it to cool, when the salt separates in small, octohedral, whitecrystals with a greenish-yellow tinge.Heated over a flame, the saltexplodes.Nitrat~urpureo-rhoa~u~n chloride, (K 0,) L( Rhp, 10NH,)Cl,, is obtainedas a pale greenish-yellow precipitate on filtering a, cold solution of thenitrate into an ice-cold mixture of 3 vols. hydrochloric acid and 1 vol.water.Nitratopurpureo-rhodium dithionate, (NO,),( Rhz,10NH3)( Sz06)2 +2H20, is prepared by precipitating a cold saturated solution of thenitrate or chloride with sodium dithionate. It forms white, silkyX-shaped aggregations resembling the corresponding cobalt salt.The air-dried salt loses all its water a t 100".It is quite insoluble i nwater. G. H. M.It is only slightly soluble in water.An aqueous solution gives all the reactions of the nitrate.Xantho-rhodium Salts. By S. M. J~SKGENSEN (J. pr. Chem. [2], 34,410-423) .-Xantho-rhodium nitrate, (NOz),( Rhs, 10NH)(3N03)4, is pre-pared by dissolving the chloropnrpureochloride in dilute sodiumhydroxide on the water-bath, and then adding pure sodium nitrite anddilute nitric acid ; the salt separates as a, white, crystalline powderwhich can be recrystallised from hot water. Heated over a flame, thINORWNIC CHEMISTRY. 115salt explodes and leaves a voluminous residue of rhodium. Thexanthonitrate is fairly soluble in cold, easily in hot water ; insolublein alcohol.Its aqueous solution, unlike that of the nitratonitrate,is not converted into the roseonitrato by boiling. It is also unactedon by sodium hydroxide, ammonium sulphide, &c.Xantho-rhodium chloride. (NO,),(Rh,,lUNH,) C14, is prepared in thesame way as the nitrate, adding dilute hydrochloric acid, however,instead of nitric acid after the sodium nitrite, when the salt separatesRS a white powder ; or from the nitrate, by filtering a concentratedsolution of the latter into dilute hydrochloric acid, the addition ofalcohol then throws down a yellowiBh-white precipitate of small octa-hedrons. When heated, the salt is decomposed, and leaves a residueof pure rhodium. Treated with silver oxide and water, the chlorideyields a solution oExantho-rhodium hydroxide, which behaves as a stronga1 kali.Xantho-rhodium platinochloride, (NO&( Rhz,10NH3) (PtC16)2, is ob-tained as a buff -colonred precipitate consisting of microscopic needlesby adding hydrogen platinochloride to a solution of the nitrate. Thissalt is analogous to the corresponding cobalt and chromium salts.Xuntho-rhodium bromide, (NOz),( Rhz,10NH3)Rr4, is obtained as awhite, crystalline preciphate on adding concentrated hydrobromic acidto a cold solution of the nitrate. I t is easily soluble in water, and thesolution gives all the reactions of the nitrate.Xardho-rhodium dithionnte, (NO,),( Rh,,lONH,)(S,O,), + 2H20, isprepared by filtering a cold solution of the nitrate into an excess ofsodium dithionate. It forms a white, crystalline precipitate consistingof microscopic prisms.Xuntho-rhodium sulphutes are prepared by treating the chloride withstrong sulphuric acid, &c. (a) The normal salt,It loses all its water at 100".(NOz) 2 ( Rh2,10NH3) ( S O&,crystallises in Bat, brilliant, needles many centimetres long. It mayalso be obtained by precipitating a solution of the acid salt withalcohol. The salt explodes on heating. With a solution of iodine inpotassium iodide, it yields a periodide.( b ) The acid salt, 2[ (NO,),(€th,,lONH,)( SO&],( H2S04),, crystallisesin long, white needles. The salt decomposes quietly on heating, leavingthe rhodium in pseudomorphs of the crystals.Xuntho-rhodium siZicoJZuoride, (NO,),( Rh,,lONH,)( SiF6),, is obtainedAS a snow-white, silky precipitate on adding a cold solution of thenitrate to an excess of hydrofluosilicic acid. Under the microscope,i t consists of small rhombic tables resembling the chloropurpureo-rhodium silicofluoride. Dilute nitric acid decomposes it, forming thenitrate and free acid.Xatztho-rhodium oxaZate, (NO,),( Rh2,10NH3) ( C,O,),, is obtained as awhite precipitate consisting of microscopic prisms, by adding a solu-tion of the nitrate to excess oE ammonium oxitlate. Nitric aciddecomposes it in the same way as the silicofluoride.The roseo- and purpureo- as well as the xantho-salts of rhodiumshow a most unmistakable resemblance to those of cobalt and chro-miurn. The xantho-rhodium salts are, however, much mom stabl116 ABSTRACTS OF CHEMICAL PAPERS.than those of cobalt and chromium. The nitrate is especially so,showing that the xantho-salts are really nitritopurpureo-salts. Thegreat stability of the nitrate is partly accounted for by the nature ofthe metal, and also probably partly by the fact that pentad nitrogenis combined with pentad nitrogen, thus :-0 : N.0