2 ABSTRACTS OF CHEMICAL PAPERS. Inorganic Chemistry. Purification of Hydrogen. By A. LIONET (Compt. rend., 89, 440). -Metallic copper removes all the impurities from hydrogen, except hydrogen phosphide. hydrogen silicide, and hydrocarbons. Cuprous oxide removes all but hydrogen silicide and the hydrocarbons. Cupric oxide removes all but the hydrocarbons. The best form of cup& oxide is that precipitated by potash from a, solution of cupric Non-existence of Nascent Hydrogen. By D. TOMNASI (Chen7. News, 40, 171) .-Reduction of Potassium Perch1ovate.--It was found that when chemically pure potassium perchlorate was submitted to the action of various reducing agents, giving nascent hydrogen, it did not undergo reduction, although it is easily transformed into chloride by the action of a compound which does not set hydrogen free, viz., sodium-hydrogen sulphite.The author asks, how can it be explained that this same perchlorate which undergoes no reduction by means of nascent hydrogen, as shown by sixteen different reactions, can be reduced by the hydrogen disengaged by the action of zinc on sodium- hydrogen sulphite. Although Wurtz declares himself to be in favour of the naacent state of bodies, it appears to the author unlikely that when hydrogen is set free by a reaction, it can be in the state of iso- lated atoms. It is known that copper, even when finely divided, is but very slightly attacked by hydrochloric acid at the ordinary tempera- ture, although copper hydride is decomposed very energetically. “ How can this fact be explained,” justly remarks Wurtz, in his Atomic Theory, “ i f to the affinity of chlorine for copper be not added the affinity of the two atoms of hydrogen to form a molecule ?” This reasoning may be said to apply equally to all the reactions producing hydrogen ; for example, we know that by the action of hydrochloric acid on zinc, there neither is nor can be any hydrogen in the state of isolated atoms, as Wurtz thinks, and the special properties of nascent hydrogen can be attributed osly to the heat which accompanies hydrogen while it is being set free.It is therefore impossible to con- clude that hydrogen can be active only in the molecular state, as hun- sulphate, and dried at 100”. C. w. w.INORGANIC CKEMISTRY. 3 dreds of examples prove to us that in many cases it is not the mole- cule of hydrogen that acts, but its atom.I n conclusion it is mentioned that the recent results of Gladstone and Tribe coincide entirely with the above hypothesis. These chemists, as is known, consider the different allotropic states of hydrogen a s ordinary hydrogen in different physical conditions. Active Condition of Oxygen induced by Nascent Hydrogen. By F. HOPPE-SEYLER (Ber., 12, 1551-1555)..-Every attempt to explain the vital processes of animals and plants necessarily implies the assumption of a cause whereby the oxygen is rendered actit-e. Hydrogen is evolved in the free state only when oxygen is not present ; and most cuTiously, in presence d oxygen, nascent hydrogen leads to energetic oxidation of any oxidisable substance which may happen to be present.The author has proved the fact by two very st'riking examples. The alloy of palla- dium with hydrogen discovered by Graham, when placed in oxygel), gives rise to water, owing to combination of the hydrogen of the alloy with the oxygen. This, of course, is well known, but it is not so well known that if indigo be present it is decolorised, and then destroyed ; that a mixture of starch withpotassium iodide is first turned blue, and. that the starch is then completely oxidised ; that ammonia is oxidised to ammonium nitrite; that benzene is oxidised to phenol; and that toluene yields benzoic acid. Perhaps a still more remarkable instance is the oxidation of rock-oil by metallic sodium in presence of the small quantity of aqueous vapour which comes in contact with it.The pro- ducts appear to be acetic and caproic acid, and perhaps butyric acid; and the hard crust which forms rmnd sodium, when it is kept under rock-oil, is really a product of oxidation of the oil, and in fact may be made to yield a number of the higher members of the fatty acid series, It thus appears to be the case that when nascent hydrogen acts on oxygen, it renders the latter gas also naseent, or a t least active. Is Ozone produced during the Atmospheric Oxidation of Phosphorus ? By C. T. KISGZETT (Chem. News, 40, 96).-It is generally believed that ozone is produced during the atmospheric oxidation of phosphorus, but the author considers it to be improbable that ozone is formed by the asrial oxidation of phosphorus, considering the constitution of ozone.Moreover, as peroxide of hydrogen is the only known agent which resembles ozone in its general properties, and it is known that hydrogen peroxide is produced in various pro- cesses of slow oxidation, it would seem likely that it is this substance which is produced in connection with the oxidation of phosphorus. In addition to various objections which the author has pointed out t o viewing the active agent produced in the atmospheric oxidation of tur- pentine as ozone, there are many considerations which lead to the con- clusion that the active agent is peroxide of hydrogen. There is no known process of slow oxidation which has been esta- blished to produce ozone. In various writings on this subject observers have always relied on properties which are common to ozone and hydrogen peroxide, and have never instituted volumetric inves- D.B. This is specially the case with ferments. W. R. b 24 ABSTRACTS OF CHEMICAL PAPERS. tigations, which are alone sufficient to decide the question. On the other hand, several processes of slow oxidation are known, in which peroxide of hydrogen is formed, as for instance, those relating to ether and the t,erpenes ; and it is thought that as hydrogen peroxide is formed in each of these cases as a secondary product, clue to the action of water on a peroxide, so also the oxidation of phosphorus by air gives rise to an oxide which generates peroxide of hydrogen by contact with water. I n conclusion, the author mentions that until it has been proved that the active agent produced in the aGrial oxidation of phosphorus has the volumetric relat'ions of ozone, such very decided statements as are to be found in chemical text-books should not be made.D. B. New Method of forming Hyponitrites and Hydroxylamine. By W. ZORN (Bey., 12, 1509--1511).-This consists in the electrolysis of a nitrite, using mercury electrodes. Thorpe describes an experiment in which he passed a current from platinum electrodes through a solu- tion of potassium nitrite, and a t the negative pole only hydrogen was evolved. On using mercury electrodes, however, if the current is stopped as soon as ammonia begins to be evolved, the liquid, after neutralisation and addition of silver nitrate, gives a copious precipitate of silver hyponitrite. I n this reduction, hydroxylamine is also formed, and it is necessary to remove it from the solution by precipitation with mercuric oxide, before adding silver nitrate to precipitate the hyponitrous acid.Four Bnnsen's elements are sufficient for t'his reac- tion; it is recommended as an advantageous method of preparing hypo ni t ri t es . W. R. Experiments tending to show the Non-elementary Charac- ter Of Phosphorus. By N, [JOCKYER (Conzpt. rend., 89, 514).- Phosphorus heated in it tube with copper gives a gas exhibiting the spectrum of hydrogen ; heated done, phosphorus gives no gas. Phos- phorus a t the negative pole of a battery in a tube-apparatus (of which a drawing is given), gives a large quantity of gas which shows the spectrum Df hydrogen, and is not phosphoretted hydrogen.c. w. w. By E. J. MAU- M S N ~ (Coinpt. rend., 89, 506).--Tn the preparation of ammonium sulphide, the hydrogen sulphide which passes through the first bottle carries ammonia, with it, and colourless crystals are deposited in the connecting tube. These crystals have the composition NH,.HS. When they are added to strong aqueous ammonia at 0", colourless crystals are deposited in a few hours having the composition (NH,),HS. The author imagines the existence of two series of ammonium- compounds containing respectively excess of ammonia and excess of hydrogen sulphide, 1 mol. of one of the constituents being united with (2, - 1) molecules of the other. Members of one series may unite with members of the other series, producing compounds like HS(NH,)15.'L[(HS),NH3] = (HS)15(NH3)1,, which might be mistaken for (NH3)HS.The compounds of ammonia with hydrochloric acid present analogies Compounds of Hydracids with Ammonia. with the above compounds. c. w. w.INORGANIC CHEMISTRY. 5 Oxygen-acids of Sulphur. By E. J. MAUXEN~ (Contpt. Tend., 89, 422).-The action of iodine on barium thiosulphate gives rise to tetrathionic acid, as observed by Fordos and Gelis, but seven other acids should be produced, according to the proportions of iodine a i d thiosulphate employed. The acids, H2S201 and H2S609, have been obtained. The latter is prepared by mixing 3 mols. of barium thio- sulphate and 2 atoms of iodine. The mixture becomes colourless in three or fonr days. It is then filtered through cotton-wool, and the crystals remaining behind are washed with alcohol.They are then pure and have the composition Bas&& ; with silver nitrate, this salt gives a white precipitate, turning black, and the liquid a t the same time becomes acid. The sodium salt crystallises in large, colourless, very soluble crystals, containing a large quantity of water. Basicity of Dithionic Acid. By H. KOLBE (J. pr. Chenz. [a], 19, 485-4,Si).-As the author hams been unable to obtain an acid salt of this acid or a neutral salt containing two bases, he doubts the cor- rectness of the usually accepted view of the bibasicity of this acid, and is now inclined to the original supposition of Berzelius that it is a monobasic acid, and is represented by the formula S0,OH. In fact, that it contains the radicle SOz, but united with only one atom of hydroxyl, that is, joined with only one atom of hydrogen by one atom of oxygen.On this supposition sulphur must exist in this acid as a pentad. That the radicle sulphoxyl (SO,) may exist as a dyad in sulphuric acid, and as a monad in dithionic acid, appears to the author to be not more improbable than the tetrad and dyad atomicities of tin in stannic and stannous compounds. Behaviour of Calcium Oxide with Carbonic Anhydride. By K. BIMBAUN and M. MAHU (Ber., 12, 1547--1561).-The object of the experiments described in this paper was to ascertain at what tempera- ture calcium oxide begins to absorb carbonic, anhydride, and a t what temperature calcium carbonate begins to dissociate. If was found that the lowest temperature at which absorption takes place is the melting point of zinc, 415-3", and that the carbonate dissociates par- tially at that temperature, although dissociation begins a t a much lower one.The amount of anhydride absorbed by the oxide is about half an equivalent. W. R. c. w. w. A. J. C. Calcium Phosphite. By R. ROTHER (Pharm. J. Trans. [3], 10, 286).--Ry adding sugar to a solution of calcium hypophosphite, the latter is precipitated, a circumstance which 'is generally unknown, and hence it is highly probable that a dense syrup of the mixed hypophos- phites contains little if any calcium salt. In the presence of iron, a, precipitate is also formed ; the proportion of sugar, however, has 110 share in this change. Ferric hypophosphite, when contained in such a sugar, is said to revert easily to th.e ferrous form, and it was found that' the ferrous salt readily oxidises even in the presence of sugar, forming the dark green and very soluble ferroso-ferric hypophosphite.Ferric hypophosphite occurs in several modifications, of which the crys-6 ABSTRACTS OF CHEMICAL PAPERS. talline variety is almost insoluble in hypophosphorous acid, and hence it is this compound which deposits from the syrup. It was attempted to regenerate this sediment by reducing it to the ferrous condition by the intervention of sulphurous acid. However, the latter was de- eomposed into sulphuric acid, sulphur, and oxygen, which reacted with the hypophosphorous acid of the sediment,, and converted it into phosphorous acid. When solutions of calcium bypophosphite and aodium sulphite are mixed, calcium sulphite is precipitated, which is redissolved by hydrochloric acid, no further reaction setting in until both the hypophosphorous and sulphurous acids are entirely freed by the addition of enough hydrochloric acid.The solution, after filtering off the sulphur, yields, on the addition of ammonia, a crystalline pre- cipitate of calcium phosphite. The latter, however, contains less than half of the phosphorous acid generated, owing to the fact that hypo- phosphorous acid is monobasic, whilst phosphorous acid is dibasic, and also that a small loss of calcium is incurred as sulphate. By treating the liquid filtered from the sulphur with calcium carbonate, a much larger amount of phosphite is thrown down than with ammonia. The addition of a solution of calcium chloride to the filtrate causes a further precipitation of phosphite, which becomes more distinctly crystalline, and subsides more rapidly when a, very little ammonia has been added to the precipitate.Calcium phosphite is a white crystalline powder, which when heated in a test-tube evolves spontaneously inflammable hydrogen phosphide accompanied by slight detonations. At a, certain temperature, it sud- denly becomes incandescent, and leaves a residue qf calcium phos- phate. D. B. Zirconium Derivatives. By S. R. PAYK~LL (Ber., 12, 1719).- The moist hydrated oxide, ZrO(OH),, absorbs carbonic anhydride from the atmosphere. By treating zirconium sulphate with the hydrate, one amorphous and two crystalline basic salts were obtained, viz., Zr02.S03+zAq, 3ZrO2.4SO3+15H,O, and 6Zr02.7S03+ 19H20.The sulphate forms with pofassium sulphate basic double salts, e.g., Researches on Erbia. By LECOQ DE BOISBAUDRAN (Cow@. rend., 89, 516).--The author examined the spectrum of erbia from various sourceEi, and with one exception the spectra thus obtained exhibited identical lines of the same intensity. The exception was the erbia derived from samarskite. The principal differences were that with samarskite-erbia, the ray in the green, X == 536.3, is much more in- tense that the ray X = 540.9, whilst in the other erbias the difference is but slight; and the line in the red, h = 640.4, is as strong, or stronger, than X = 653.4; whereas i n the other erbias, the line h = 653.4 is much stronger than k = 640.4.Two specimens of erbia were taken, one nearly pure, giving the normal spectrum, the other from samarskite, and containing a large quantity of yttria. On fractionation by means of ammonia and sub- sequently by potassium or sodium sulphate, a portion was obtained from the first which gave a spcctrum resembling that of the second, K,0~60,+2(12~.0.02.50~) + 1%H,O. w. c. w.INORGANIC CHEMISTRY. 7 and a portion was obtained from the second giving ft spectrum like that, of the first. The author is continuing this research. c. w. w. Two New Elements in Erbia. By P. T. CLEVE (Compt. rend., 89, 478).-In attempting to obtain pure erbia, the author was led to suspect the existence of two other earths in the erbia obtained. The mixture was therefore fractionated, and the different fractions ex- amined spectroscopically.It was found that, in addition t o bands common to all, one band X = 6840 was strong in the residues rich in ytterbia, and want’ing in those containing yttria and erbia, whilst two others, X = 6400 and 5360, were strong in the yttria and ytterbia residues. The colour of the fraction treated for ytterbia was a violet-rose, whilst the yttria fraction had an orange tint. The metal characterised by the first band, X = 6840, the author pro- poses to name thuZium ; it would have an atomic weight of about 113 (the oxide being TmO). Pure erbia, to which must be attributed the common bands, has probably an atomic weight of 110-111. Its oxide has a clear rose colonr. The third metal, holmium, is charncterised by the bands X = 6400 and 5360 ; it should have an atomic weight less than 108 ; its oxide seems t o be yellow.c. w. w. Spectra of the Earths of the Yttria-group. By J. L. SORET (Compt. rend., 89, 521).-The author considers that the new earth, ldmiu, discovered by ClBve, is identical with an earth discovered by Delafontaine and Marignac, whose absorption-spectrum was described by himself (Compt. ren,d., April 29, 1878), and to which Delafontaine gave the name plziZ@pia. CIBre’s holmium is characterised by two bands X = 640 and X = 536, and these two belong likewise t o philip- pium, which is characterised by several other bands. Clhe’s second earth, thdia, is characterised by a red ray X = 684. The author claims to have indicated the existence of this element also ( A ~ c h .Sci., 63, 99). Marignac also showed the probable existence of this earth in the products rich in philippia and having a low equiva- lent. c. w. w. Scandium. By P. T. CLEVE (Compt. rend., 89, 419).-This metal occurs only in gadolinite (0.002 to 0.003 per cent.) and yttrotitanite (0.005 per cent.). Scandium forms but one oxide, scandia, Sc,03; the composition of which is proved by that of potassium scandium oxalate, and of the double sulphates of scandium with the sulphates of potas- sium and with ammonium. 8 to 10 grams of scandinm oxide, having a molecular weight of 106, agreeing with the number obtained by Nilson, yielded, by repeated decompositions of its nitrate, about 1 gram of a white oxide. This was converted into sulphate, and 1.451 grams of this sulphate yielded 0.5293 gram of scandium oxide.The atomic weight of the metal is therefore 44.91, and the mole- cular weight of the oxide, considered as ScO, is 45.94 (? 60.91 = 44.91 + 16), diflering greatly from the lowest number found by Nilson, viz., 105.83. The author considers that this is due to a large8 ABSTRACTS OF CHEMICAL PAPERS. admixture of ytterhia in Nilson’s scandia. The atomic weight, as determined by the decomposition of the nitrate, is 45.12. The atomic weight of scandium may therefore be takeu as 45. Scandizim oxide o r Scandin, Sc203? is a light, white, infusible powder of sp. gr. 3.8, resembling magnesia; it is scarcely soluble even in strong acids, but more so than alumina. Sulphuric acid converts it into a bulky white mass of sulphate ; hydrochloric acid dissolves the oxide more easily than nit,& acid.Scandium hydrate is a bulky white precipitate, drying up to semi- transparent fragments. It does not absorb carbonic acid from the air, is insoluble in ammonia or in potash, and does not decompose ammonium salts when heated with them. Scandium salts are colourless or white, and have an acid, astringent taste, very different from the sweet taste of the salts of the yttrium metals. The sulphate does not form distinct crystals; the nitrate, oxalate, acetate, and formate, are crystallisab!e. The chloride ex- hibits the following reactions :-It gives no spectrum when heated in it gas flame. Potash and ammonia produce bulky white precipitates, insoluble in excess ; tartaric acid prevents the precipitation by ammonia in the cold, but on heating an abundant precipitate falls.Sodium carbonate gives a precipitate, soluble in excess. Sulphuretted hydrogen produces no change ; ammonium sulphide precipitates the hydrate. Sodium orthophosphate gives a gelatinous precipitate. Oxalic acid give8 a curdy precipitate, quickly becoming crystalline ; this preci- pitate dissolves in strong acids, and cannot be completely reprecipitated. Although it appears more soluble than the oxalates of the other yttrium metals, it is found in the first precipitates in the fractional precipitation of a mixture of scandium and ytterbium by oxalic acid. Acid potassium oxalate precipitates a crystalline double salt. Sodium hyposulphite precipitates a boiling solution easily, but incompletely.Sodium acetate behaves similarly. The sulphates of potassium and sodium precipitate crystalline double salts, soluble in a saturated solution of the precipitant. The author describesin a previous paper (BUZZ. SOC. Chinz., 31, 486) the chloride, nitrate, and sulphate of scandium ; the double sulphates, the double oxalate, Sc,(C,O,),.K2C2O4.3H,O ; the acetate, the formate, and selenite, 3Sc,0,.10SeO2.4H2O. The existence of scandium was predicted by Mendelejeff, and its pro- perties described under the name of ekabor (Anmalen, Sq., 8, 1.13). The following table shows a comparison of the observed properties of scandium witlh those predicted of ekabor. SC,(SO~),.~K$~O~, S C , ( S O ~ ) , . ~ N ~ S O ~ . ~ ~ H ~ O , S C ~ ( SOi)s.(NH4)2SO4 ; 8upposed Characters of Ekabor.Atomic Weight = 44. Ekabor should have but one stable oxide, Ebz03, :t stronger base than alumina, which it should resemble in mauy respects. It should be less basic than mag- nesia. Observed Ch,aracters of Scandium. Atomic Weight = 45. Scandium forms only one oxide, Sc203, more energetic than alu- mina, and less so than magnesia.INORGANIC CHEMISTRY. 9 Ekabor oxide should resemble yttria, although less basic. The se- paration of these two earthswill be difficult, depending on differences of solubility or of basicity. Oxide of ekabor is insoluble in alkalis; it will probably not de- compose ammonium salts. The salts should be colourless, and give gelatinous preci.pitates with KHO, Na2C03, and HNaS02. The sulphate should form a double salt with K2S04, having the composition of alum, but not isomorphous with it.But few ekabor salts should crystallise well. The anhydrous chloride should be decomposed by water, giving off hydrochloric acid. The oxide is infusible a,nd so- luble with difficulty in acids after ignition. The density of the oxide would be about 3.5. Scnndia is less basic than yttria, and their separation depends on differences of solubility between their nitrates. Scandium hydrate is insoluble in alkalis ; it does not decompose ammonium chloride. The salts are colourless, and give gelatinous precipitates with KHO, Na2C03, and HNaS02. Potassium - scandium sulphate is anhydrous, but otherwise cor- responds in composition with alum. Scandium sulphate does not form distinct crystals, but the nitratc, acetate, and formatme crys- tallise well. The crystallised chloride is de- composed by heat, giving off hydrochloric acid.The oxide is an infusible powder, nearly insoluble in acids after ignition. The density of the oxide = 3.8. c. w. w. Absorption of Nitrogen Dioxide by Ferrous Salts. By J. GAY (Compt. rend., 89, 410).-Peligot assigned the formula 4FeS04.N20, to the compound of nitrogen dioxide with ferrous su1- phate. The author finds that the composition of this body depends on t'he temperature and on the pressurc of the residual nitrogen dioxide. At temperatures up to 8" and at the ordinary pressure, the com- pound formed has the formula 3FeS01.N202 ; from 8" to 25", at the atmospheric pressure, i t has the formuh 4FeSO4.N2O2 ; at tempera- tures above 25" nitrogen dioxide is rapidly given off, and the com- pound 5FeSO4.N,O, is produced.All these compounds exhibit very marked tensions of dissociation, a fact which explains their decomposition in st vacuum ; they are also decomposed by a current of hydrogen. Reducing agents, such as ferrous oxide, reduce the nitrogen dioxide, a mixture of monoxide and free nitrogen being evolved, while the temperature rises sensibly. c. w. w. Nitrosothioferrates. By J. 0. R l ~ ~ ~ (Ber. 12,1715-1717). -By the action of potassium nitrite and ammonium sulphide on a ferrous salb, Roussin (Ann. Chim. Phys. [3], 52, 285) obtained a black substance, which was afterwards examined by Porczinsky (Annden, 125, 302), Demel (Be?-., 12, 46l), and Pawel (ihid., 12, 1407). This is named by the author ammoiziunz nitrosoferrclthiofen.ate.10 ABSTRACTS OF CHEMICAL PAPERS.On the addition of an alkali ferrous oxide is precipitated, and potassium nitrosothioferrate is obtained. The free acid which is liberated when this salt is treated with hydrochloric acid, combines with alkaline sulphides to form a red salt, to which the name nitroso- ferrous potassimn sulphide is given. Salts correspondiug with each of the two first-mentioned acids have been prepared. They are all converted into nitroprussides by the action of potassium cyanide. w. c. w. Potassium and Ammonium Ferric Chromates. By C. HENSGEN (Rer., 12,1656-1658) .-These salts separate out in dark-red plates containing 4 mols. H20 [K or NH,], when a solution containing ferric chloride and ammonium or potassium dichromate, is slowly evaporated.They have the formula KzCr04.Fe2(Cr04)3.4H,0. The ammonium salt is decomposed by cold water and also by the action of heat. w. c. w. Contributions to the Chemistry of the Chrornammonium- compounds. By S. 31. JORGENSER (J. p r . Chem. [el, 20, 105-145. -1. C l ~ l o r o p u ~ p u ~ e o - c l ~ ~ o ~ ~ a i ~ ~ ~ ~ ~ Salts.-The starting point for these salts is the chloride, CI,(Cr210NH3) C1,. This is prepared by reducing violet chromic chloride in a stream of pure dry hydrogen, at a red heat, and adding it to a solution of ammonium chloride in strong ammonia (25 grams Cr,CI, reduced to Cr2C14, 90 grams NH4C1, 0.5 litre ammonia). Air is then passed through the blue liquid until oxidation is complete. Two litres of crude hydrochloric acid are added, and the mixture is boiled for some minutes, during which chloropurpureo-chromium chloride separates as a carmine-coloured powder.The crude chloride is washed with a mixture of equal volumes of hydrochloric acid and water, dissolved in very weak sulphuric acid, and filtered into a great excess of strong cold hydrochloric acid. The resulting precipitate is boiled with hydrochloric acid, and washed first with a mixture of acid and water, then with alcohol, and finally dried in the air a t the ordinary temperature. This chloride is also a bye-product in preparing ClAve’s tetramine chloride by the following process :-Ammonium dichromate is I educed by boiling with hydrochloric acid and alcohol, and after addition of ammonium chloride the liquid is evaporated to dryness.The dry residue is then dissolved in strong ammonia ; strong hydrochloric acid is added, and the crystals which are deposited on standing are washed first with a mixture of equal parts of hydrochloric acid and water until free from ammonium chloride, then with water, and finally dried. It consists of a mixture of chromium-tetramine chloride and chloropnrpureo chloride. This mixture must be protected from the action of light during the remaining operations. It is dissolved in cold water, and shaken with a solution of one part of ammonium sulphate in five parts of water. The tetramine chlorosulphate precipitates in crystals ; the filtrate containing the purpureochloride is mixed with hydrosilico- fluoric acid, and gives a precipitate of chloropurpureo-chromium silico- fluoride.After being washed, it is treated with dilute hydrochloricIN ORGANIC CHEMISTRY. 11 acid, to reconvert it into chloride ; after reprecipitation with strong acid and washing, first with dilute acid and then with alcohol, it is quite pure. The two salts may also be separated by taking advantage of the insolubility of the componnd C1,( CrJONH,) ( HgsC1,),, produced by adding mercuric chloride to the mixture. The mercury-compound atter washing is easily reconverted into the chloride by treatment with h-j-drochloric acid. Chloropurpureo-chromium chioride is a red crystalline powder, of a purer red colour than the corresponding cobalt-compound. It appears to crystallise in octohedra of sp. gr. 1.687. It dissolves in 154 parts of water a t 16", and forms a violet-red solution, which, on exposure to light, deposits chromium hydrate.When it is kept, even in the dark, or boiled, roseochromium chloride is produced. It gives the follow- ing reactions :-With sodium hypochlorite, nitrogen is evolved, and the chromium is oxidised to chromic acid. Its solution gives a preci- pitate with strong hydrochloric acid, owing to the insolubility of the chloride in acid. With hydrobromic acid, it gives a crystalline preci- pitate of the bromide, and with solid potassium iodide one of the iodide. When boiled with potassium cyanide, it turns yellow. Strong nitric acid precipitates the chloro-nitrate. Hydrosilicofluoric acid throws down the red crystalline chlorosilicofluoride. Platinic chloride precipitates, even from a very dilute solution, the chloropurpureo- chromium platinochloride.Sodium platino-bromide gives an ana- logous precipitate. Mercuric chloride gives red needles of the double salt. Precipitates are also produced by potassium mercuribromide and iodide, by sodium dithionate, potassium chromate, and dichromate, ammonium molybdate, and phosphomolybdate, and by picric and oxalic acids. I n these respects this salt closely resembles the analogous cobalt salt. On treatment with silver nitrate only four atoms of chlorine are removed, and the chloro-nitrate is formed. By rubbing the solid chloride with silver oxide and water, raseochromium hy- drate is formed. It is a, deep red alkaline liquid, which gives a yellowish-red precipitate of roseochromium bromide with strong hydrobromic acid ; this, when boiled with hydrobromic acid, changes to bromopurpureo-chromium bromide.In the chloro-chloride, the radicle chlorine is so firmly combined that hot strong sulphuric acid does not expel i t ; the product is acid chloro-sulphate. Towards reducing agents, however, the chromium series differ in behavionr from the cobalt series, for the chromium is not so easily re- duced. With snlphuretted hydrogen, or with ammonium sulphide, the purpureo-cobalt-compounds give cobalt sulphide, but the purpui-eo- chromium compounds suffer no change, except the forniation of a crystalline pnrpureopolysulphide, if the ammonium sidphide contains much free sulphur. The cobalt salts are also reduced by potassium ferrocyanide, whereas the chromium salts give a precipitate of ferro- cyanide of chloropurpureo-chromium.The latter part of this paper is occupied with detailed descriptions and analyses of numerous salts of chloropurpureo-chromium chloride, prepared by double decomposition. They have all a red or orange-red colour, and closely resemble the corresponding salts of chloropurpureo- cobalt. W. R.12 ABSTRACTS OF CHEMICAL PAPERS. Behaviour of Copper- Ammonium Chloride with Ferrous Sulphide. By W. F. K. STOCK (C'hern. News, 40, 159).--In the course of recent experiments on the accurate determination of carbon in iron and steel containing much sulphur, i t appeared desirable to ascertain definitely in what manncr the reagent used for the carbon separation acted on the iron sulphnr compound, but as the composition of that compound is unknown, it was thought best t80 experiment with a sulphide of known quality.The process used for the carbon separa- tion was McCreath's method based on treating a weighed quantity of iron or steel with a hot concentrated solution of copper ammonium chloride. From the results, it was evident that the actions of the double cop- per ammonium salt upon iron carbide and upon iron sulphide were pre- cisely analogous, and that tbe method held out no hope of separation. It only remained to find to what extent the decomposition had pro- ceeded during the exposure, which was for half an hour a t nearly boiling heat, It is shown that allowing for oxidation during washing, &c., it may safely be assumed that 80 per cent. of the original sul- phide was decomposed by the double copper salt with liberation of the corresponding amount of free sulphur. A second experiment was made with native ferric sulphide, which was very finelypowdered and exposed a t boiling heat for over an hour t20 the copper solution, but although some free sulphur was obtained, the deconiposition was far from complete.Action of the HaloTd Acids on the Sulphates of Mercury. By A. DITTE (Ann,. Qhim. Phys. [ 5 ] , l7,120--128).-1t has beenstated that dry hydrochloiic acid gas decomposes mercuric sulphate, forming mercuric chloride and free sulphurie acid, and t h a t since the chloride is more volatile than the acid, the former can be separated by sublima- tion at a suitable temperature ; and further, that hydriodic and hydro- cyanic acids act in a similar manner. The author shows that these statements are wholly incorrect.When dry hydrochloric acid is passed over mercuric sulphnte a t ordinary temperatures, no reaction ensues : on warming the sulphate, absorption takes place, with disengagement of heat and without forma- tion of water ; on heating more strongly, the product sublimes, but the crystals hare no resemblance whatever to sublimed mercuric chlo- ride. An analysis of the crjstals showed that their composition exactly corresponded with the formula HgSO, 2HCl; they are very hygrometric, dissolving in water apparently without decomposition. When volatilised they do not disengage hydrochloric acid. Hydrobromic acid gas acts in a precisely similar manner, forming the compound HgS01.2HBr.The body HgSO,.BHCl is likewise formed with great facility by gently heating a, mixture of sulphuric acid and mercuric chloride, in molecular proportions ; or by dissolving the neutral sulphate in hydro- chloric acid, and evaporating until crystals are obtained. The action of hydriodic acid is different; sulphuric acid decom- poses mercuric iodide on heatir,g, no compound o f thc formula €€gSOa.2HI being formed. In the same manner, solution of hydriodic D. B.MINERALOGICAL CHEMISTRY. 13 acid in excess, partly or wholly decomposes mercuric sulphate, but no definite combination takes place. Hydrofluoric and hydrocyanic acids are without action on mercuric sulphate. Basic mercuric sulphate, twpeth wahzeral, acts with regard to hydro- chloric acid in a manner analogous to mercuric sulphate, but it ab- sorbs 6 molecules of HCl for every molecule of sulphate, forming the compound HgS04.2Hg0.6HC1 ; the latter on being heated strongly, breaks up into the mercuric compound and mercuric chloride, Hg3S06.6HC1 = HgS04.2HCl + 2HgC12 + ‘2H20. thus :.- A precisely similar compound is formed by the action of either gaseous or liquid hydrobromic acid on tnrpeth mineral.A New Salt of an Iridammonium. By K. BIRNBAUV (Bey., 12, 1544--1547).-By boiling the double salt of iridic sulphite and sodium sulphite with hydrochloric acid, an acid salt is formed pre- sumably of the formula Irz( S03)3.3NaHS03. When its solution is saturated with gaseous ammonia, a compound crystallises out in red crusts, having the formula Ir2Naa( S03)G(NH3),.10H20.The author assigns to it the constitutional formula- SO3 : Ir(NH3)3 NH4 Na so,< 1 SO3 : Ir(N%)s J. W. + 3 >s03.0H2c), and supposes the SO3 group t o be i n combination with an irid- ammonium of the formula (NH&Ir2. W. R.2 ABSTRACTS OF CHEMICAL PAPERS.Inorganic Chemistry.Purification of Hydrogen. By A. LIONET (Compt. rend., 89,440). -Metallic copper removes all the impurities from hydrogen,except hydrogen phosphide. hydrogen silicide, and hydrocarbons.Cuprous oxide removes all but hydrogen silicide and the hydrocarbons.Cupric oxide removes all but the hydrocarbons. The best form ofcup& oxide is that precipitated by potash from a, solution of cupricNon-existence of Nascent Hydrogen. By D. TOMNASI (Chen7.News, 40, 171) .-Reduction of Potassium Perch1ovate.--It was foundthat when chemically pure potassium perchlorate was submitted to theaction of various reducing agents, giving nascent hydrogen, it didnot undergo reduction, although it is easily transformed into chlorideby the action of a compound which does not set hydrogen free, viz.,sodium-hydrogen sulphite.The author asks, how can it be explainedthat this same perchlorate which undergoes no reduction by means ofnascent hydrogen, as shown by sixteen different reactions, can bereduced by the hydrogen disengaged by the action of zinc on sodium-hydrogen sulphite. Although Wurtz declares himself to be in favourof the naacent state of bodies, it appears to the author unlikely thatwhen hydrogen is set free by a reaction, it can be in the state of iso-lated atoms.It is known that copper, even when finely divided, is butvery slightly attacked by hydrochloric acid at the ordinary tempera-ture, although copper hydride is decomposed very energetically.“ How can this fact be explained,” justly remarks Wurtz, in his AtomicTheory, “ i f to the affinity of chlorine for copper be not added theaffinity of the two atoms of hydrogen to form a molecule ?” Thisreasoning may be said to apply equally to all the reactions producinghydrogen ; for example, we know that by the action of hydrochloricacid on zinc, there neither is nor can be any hydrogen in the state ofisolated atoms, as Wurtz thinks, and the special properties of nascenthydrogen can be attributed osly to the heat which accompanieshydrogen while it is being set free.It is therefore impossible to con-clude that hydrogen can be active only in the molecular state, as hun-sulphate, and dried at 100”. C. w. wINORGANIC CKEMISTRY. 3dreds of examples prove to us that in many cases it is not the mole-cule of hydrogen that acts, but its atom.I n conclusion it is mentioned that the recent results of Gladstoneand Tribe coincide entirely with the above hypothesis. These chemists,as is known, consider the different allotropic states of hydrogen a sordinary hydrogen in different physical conditions.Active Condition of Oxygen induced by Nascent Hydrogen.By F. HOPPE-SEYLER (Ber., 12, 1551-1555)..-Every attempt toexplain the vital processes of animals and plants necessarily impliesthe assumption of a cause whereby the oxygen is rendered actit-e.Hydrogen is evolved in the free state only when oxygen is not present ;and most cuTiously, in presence d oxygen, nascent hydrogen leads toenergetic oxidation of any oxidisable substance which may happen tobe present. The author hasproved the fact by two very st'riking examples.The alloy of palla-dium with hydrogen discovered by Graham, when placed in oxygel),gives rise to water, owing to combination of the hydrogen of the alloywith the oxygen. This, of course, is well known, but it is not so wellknown that if indigo be present it is decolorised, and then destroyed ;that a mixture of starch withpotassium iodide is first turned blue, and.that the starch is then completely oxidised ; that ammonia is oxidisedto ammonium nitrite; that benzene is oxidised to phenol; and thattoluene yields benzoic acid.Perhaps a still more remarkable instanceis the oxidation of rock-oil by metallic sodium in presence of the smallquantity of aqueous vapour which comes in contact with it. The pro-ducts appear to be acetic and caproic acid, and perhaps butyric acid;and the hard crust which forms rmnd sodium, when it is kept underrock-oil, is really a product of oxidation of the oil, and in fact may bemade to yield a number of the higher members of the fatty acid series,It thus appears to be the case that when nascent hydrogen acts onoxygen, it renders the latter gas also naseent, or a t least active.Is Ozone produced during the Atmospheric Oxidation ofPhosphorus ? By C.T. KISGZETT (Chem. News, 40, 96).-It isgenerally believed that ozone is produced during the atmosphericoxidation of phosphorus, but the author considers it to be improbablethat ozone is formed by the asrial oxidation of phosphorus, consideringthe constitution of ozone. Moreover, as peroxide of hydrogen is theonly known agent which resembles ozone in its general properties,and it is known that hydrogen peroxide is produced in various pro-cesses of slow oxidation, it would seem likely that it is this substancewhich is produced in connection with the oxidation of phosphorus. Inaddition to various objections which the author has pointed out t oviewing the active agent produced in the atmospheric oxidation of tur-pentine as ozone, there are many considerations which lead to the con-clusion that the active agent is peroxide of hydrogen.There is no known process of slow oxidation which has been esta-blished to produce ozone.In various writings on this subjectobservers have always relied on properties which are common to ozoneand hydrogen peroxide, and have never instituted volumetric inves-D. B.This is specially the case with ferments.W. R.b 4 ABSTRACTS OF CHEMICAL PAPERS.tigations, which are alone sufficient to decide the question. On theother hand, several processes of slow oxidation are known, in whichperoxide of hydrogen is formed, as for instance, those relating to etherand the t,erpenes ; and it is thought that as hydrogen peroxide is formedin each of these cases as a secondary product, clue to the action of wateron a peroxide, so also the oxidation of phosphorus by air gives rise toan oxide which generates peroxide of hydrogen by contact with water.I n conclusion, the author mentions that until it has been provedthat the active agent produced in the aGrial oxidation of phosphorushas the volumetric relat'ions of ozone, such very decided statements asare to be found in chemical text-books should not be made.D. B.New Method of forming Hyponitrites and Hydroxylamine.By W. ZORN (Bey., 12, 1509--1511).-This consists in the electrolysisof a nitrite, using mercury electrodes. Thorpe describes an experimentin which he passed a current from platinum electrodes through a solu-tion of potassium nitrite, and a t the negative pole only hydrogen wasevolved.On using mercury electrodes, however, if the current isstopped as soon as ammonia begins to be evolved, the liquid, afterneutralisation and addition of silver nitrate, gives a copious precipitateof silver hyponitrite. I n this reduction, hydroxylamine is also formed,and it is necessary to remove it from the solution by precipitationwith mercuric oxide, before adding silver nitrate to precipitate thehyponitrous acid. Four Bnnsen's elements are sufficient for t'his reac-tion; it is recommended as an advantageous method of preparinghypo ni t ri t es . W. R.Experiments tending to show the Non-elementary Charac-ter Of Phosphorus.By N, [JOCKYER (Conzpt. rend., 89, 514).-Phosphorus heated in it tube with copper gives a gas exhibiting thespectrum of hydrogen ; heated done, phosphorus gives no gas. Phos-phorus a t the negative pole of a battery in a tube-apparatus (of whicha drawing is given), gives a large quantity of gas which shows thespectrum Df hydrogen, and is not phosphoretted hydrogen. c. w. w.By E. J. MAU-M S N ~ (Coinpt. rend., 89, 506).--Tn the preparation of ammoniumsulphide, the hydrogen sulphide which passes through the first bottlecarries ammonia, with it, and colourless crystals are deposited in theconnecting tube. These crystals have the composition NH,.HS. Whenthey are added to strong aqueous ammonia at 0", colourless crystalsare deposited in a few hours having the composition (NH,),HS.The author imagines the existence of two series of ammonium-compounds containing respectively excess of ammonia and excess ofhydrogen sulphide, 1 mol.of one of the constituents being unitedwith (2, - 1) molecules of the other. Members of one series mayunite with members of the other series, producing compounds likeHS(NH,)15.'L[(HS),NH3] = (HS)15(NH3)1,, which might be mistakenfor (NH3)HS.The compounds of ammonia with hydrochloric acid present analogiesCompounds of Hydracids with Ammonia.with the above compounds. c. w. wINORGANIC CHEMISTRY. 5Oxygen-acids of Sulphur. By E. J. MAUXEN~ (Contpt. Tend.,89, 422).-The action of iodine on barium thiosulphate gives rise totetrathionic acid, as observed by Fordos and Gelis, but seven otheracids should be produced, according to the proportions of iodine a i dthiosulphate employed.The acids, H2S201 and H2S609, have beenobtained. The latter is prepared by mixing 3 mols. of barium thio-sulphate and 2 atoms of iodine. The mixture becomes colourless inthree or fonr days. It is then filtered through cotton-wool, and thecrystals remaining behind are washed with alcohol. They are thenpure and have the composition Bas&& ; with silver nitrate, this saltgives a white precipitate, turning black, and the liquid a t the sametime becomes acid. The sodium salt crystallises in large, colourless,very soluble crystals, containing a large quantity of water.Basicity of Dithionic Acid. By H. KOLBE (J. pr. Chenz. [a],19, 485-4,Si).-As the author hams been unable to obtain an acid saltof this acid or a neutral salt containing two bases, he doubts the cor-rectness of the usually accepted view of the bibasicity of this acid,and is now inclined to the original supposition of Berzelius that it isa monobasic acid, and is represented by the formula S0,OH.In fact,that it contains the radicle SOz, but united with only one atom ofhydroxyl, that is, joined with only one atom of hydrogen by one atomof oxygen. On this supposition sulphur must exist in this acid as apentad.That the radicle sulphoxyl (SO,) may exist as a dyad in sulphuricacid, and as a monad in dithionic acid, appears to the author to be notmore improbable than the tetrad and dyad atomicities of tin in stannicand stannous compounds.Behaviour of Calcium Oxide with Carbonic Anhydride.ByK. BIMBAUN and M. MAHU (Ber., 12, 1547--1561).-The object of theexperiments described in this paper was to ascertain at what tempera-ture calcium oxide begins to absorb carbonic, anhydride, and a t whattemperature calcium carbonate begins to dissociate. If was foundthat the lowest temperature at which absorption takes place is themelting point of zinc, 415-3", and that the carbonate dissociates par-tially at that temperature, although dissociation begins a t a muchlower one. The amount of anhydride absorbed by the oxide is abouthalf an equivalent. W. R.c. w. w.A. J. C.Calcium Phosphite. By R. ROTHER (Pharm. J. Trans. [3], 10,286).--Ry adding sugar to a solution of calcium hypophosphite, thelatter is precipitated, a circumstance which 'is generally unknown, andhence it is highly probable that a dense syrup of the mixed hypophos-phites contains little if any calcium salt.In the presence of iron, a,precipitate is also formed ; the proportion of sugar, however, has 110share in this change. Ferric hypophosphite, when contained in sucha sugar, is said to revert easily to th.e ferrous form, and it was foundthat' the ferrous salt readily oxidises even in the presence of sugar,forming the dark green and very soluble ferroso-ferric hypophosphite.Ferric hypophosphite occurs in several modifications, of which the crys6 ABSTRACTS OF CHEMICAL PAPERS.talline variety is almost insoluble in hypophosphorous acid, and henceit is this compound which deposits from the syrup.It was attemptedto regenerate this sediment by reducing it to the ferrous condition bythe intervention of sulphurous acid. However, the latter was de-eomposed into sulphuric acid, sulphur, and oxygen, which reactedwith the hypophosphorous acid of the sediment,, and converted itinto phosphorous acid. When solutions of calcium bypophosphite andaodium sulphite are mixed, calcium sulphite is precipitated, which isredissolved by hydrochloric acid, no further reaction setting in untilboth the hypophosphorous and sulphurous acids are entirely freed bythe addition of enough hydrochloric acid. The solution, after filteringoff the sulphur, yields, on the addition of ammonia, a crystalline pre-cipitate of calcium phosphite.The latter, however, contains less thanhalf of the phosphorous acid generated, owing to the fact that hypo-phosphorous acid is monobasic, whilst phosphorous acid is dibasic, andalso that a small loss of calcium is incurred as sulphate. By treatingthe liquid filtered from the sulphur with calcium carbonate, a muchlarger amount of phosphite is thrown down than with ammonia. Theaddition of a solution of calcium chloride to the filtrate causes a furtherprecipitation of phosphite, which becomes more distinctly crystalline,and subsides more rapidly when a, very little ammonia has been addedto the precipitate.Calcium phosphite is a white crystalline powder, which when heatedin a test-tube evolves spontaneously inflammable hydrogen phosphideaccompanied by slight detonations.At a, certain temperature, it sud-denly becomes incandescent, and leaves a residue qf calcium phos-phate. D. B.Zirconium Derivatives. By S. R. PAYK~LL (Ber., 12, 1719).-The moist hydrated oxide, ZrO(OH),, absorbs carbonic anhydridefrom the atmosphere. By treating zirconium sulphate with thehydrate, one amorphous and two crystalline basic salts were obtained,viz., Zr02.S03+zAq, 3ZrO2.4SO3+15H,O, and 6Zr02.7S03+ 19H20.The sulphate forms with pofassium sulphate basic double salts, e.g.,Researches on Erbia. By LECOQ DE BOISBAUDRAN (Cow@. rend.,89, 516).--The author examined the spectrum of erbia from varioussourceEi, and with one exception the spectra thus obtained exhibitedidentical lines of the same intensity.The exception was the erbiaderived from samarskite. The principal differences were that withsamarskite-erbia, the ray in the green, X == 536.3, is much more in-tense that the ray X = 540.9, whilst in the other erbias the differenceis but slight; and the line in the red, h = 640.4, is as strong, orstronger, than X = 653.4; whereas i n the other erbias, the lineh = 653.4 is much stronger than k = 640.4.Two specimens of erbia were taken, one nearly pure, giving thenormal spectrum, the other from samarskite, and containing a largequantity of yttria. On fractionation by means of ammonia and sub-sequently by potassium or sodium sulphate, a portion was obtainedfrom the first which gave a spcctrum resembling that of the second,K,0~60,+2(12~.0.02.50~) + 1%H,O.w. c. wINORGANIC CHEMISTRY. 7and a portion was obtained from the second giving ft spectrum likethat, of the first.The author is continuing this research. c. w. w.Two New Elements in Erbia. By P. T. CLEVE (Compt. rend.,89, 478).-In attempting to obtain pure erbia, the author was led tosuspect the existence of two other earths in the erbia obtained. Themixture was therefore fractionated, and the different fractions ex-amined spectroscopically. It was found that, in addition t o bandscommon to all, one band X = 6840 was strong in the residues rich inytterbia, and want’ing in those containing yttria and erbia, whilst twoothers, X = 6400 and 5360, were strong in the yttria and ytterbiaresidues.The colour of the fraction treated for ytterbia was a violet-rose,whilst the yttria fraction had an orange tint.The metal characterised by the first band, X = 6840, the author pro-poses to name thuZium ; it would have an atomic weight of about 113(the oxide being TmO).Pure erbia, to which must be attributed thecommon bands, has probably an atomic weight of 110-111. Its oxidehas a clear rose colonr. The third metal, holmium, is charncterised bythe bands X = 6400 and 5360 ; it should have an atomic weight lessthan 108 ; its oxide seems t o be yellow. c. w. w.Spectra of the Earths of the Yttria-group. By J. L. SORET(Compt. rend., 89, 521).-The author considers that the new earth,ldmiu, discovered by ClBve, is identical with an earth discovered byDelafontaine and Marignac, whose absorption-spectrum was describedby himself (Compt.ren,d., April 29, 1878), and to which Delafontainegave the name plziZ@pia. CIBre’s holmium is characterised by twobands X = 640 and X = 536, and these two belong likewise t o philip-pium, which is characterised by several other bands.Clhe’s second earth, thdia, is characterised by a red ray X = 684.The author claims to have indicated the existence of this element also( A ~ c h . Sci., 63, 99). Marignac also showed the probable existence ofthis earth in the products rich in philippia and having a low equiva-lent. c. w. w.Scandium. By P. T. CLEVE (Compt. rend., 89, 419).-This metaloccurs only in gadolinite (0.002 to 0.003 per cent.) and yttrotitanite(0.005 per cent.).Scandium forms but one oxide, scandia, Sc,03; thecomposition of which is proved by that of potassium scandium oxalate,and of the double sulphates of scandium with the sulphates of potas-sium and with ammonium. 8 to 10 grams of scandinm oxide, havinga molecular weight of 106, agreeing with the number obtained byNilson, yielded, by repeated decompositions of its nitrate, about 1 gramof a white oxide. This was converted into sulphate, and 1.451 gramsof this sulphate yielded 0.5293 gram of scandium oxide.The atomic weight of the metal is therefore 44.91, and the mole-cular weight of the oxide, considered as ScO, is 45.94 (? 60.91 =44.91 + 16), diflering greatly from the lowest number found byNilson, viz., 105.83. The author considers that this is due to a larg8 ABSTRACTS OF CHEMICAL PAPERS.admixture of ytterhia in Nilson’s scandia.The atomic weight, asdetermined by the decomposition of the nitrate, is 45.12. The atomicweight of scandium may therefore be takeu as 45.Scandizim oxide o r Scandin, Sc203? is a light, white, infusible powderof sp. gr. 3.8, resembling magnesia; it is scarcely soluble even instrong acids, but more so than alumina. Sulphuric acid converts itinto a bulky white mass of sulphate ; hydrochloric acid dissolves theoxide more easily than nit,& acid.Scandium hydrate is a bulky white precipitate, drying up to semi-transparent fragments. It does not absorb carbonic acid from theair, is insoluble in ammonia or in potash, and does not decomposeammonium salts when heated with them.Scandium salts are colourless or white, and have an acid, astringenttaste, very different from the sweet taste of the salts of the yttriummetals.The sulphate does not form distinct crystals; the nitrate,oxalate, acetate, and formate, are crystallisab!e. The chloride ex-hibits the following reactions :-It gives no spectrum when heated init gas flame. Potash and ammonia produce bulky white precipitates,insoluble in excess ; tartaric acid prevents the precipitation byammonia in the cold, but on heating an abundant precipitate falls.Sodium carbonate gives a precipitate, soluble in excess. Sulphurettedhydrogen produces no change ; ammonium sulphide precipitates thehydrate. Sodium orthophosphate gives a gelatinous precipitate.Oxalicacid give8 a curdy precipitate, quickly becoming crystalline ; this preci-pitate dissolves in strong acids, and cannot be completely reprecipitated.Although it appears more soluble than the oxalates of the otheryttrium metals, it is found in the first precipitates in the fractionalprecipitation of a mixture of scandium and ytterbium by oxalic acid.Acid potassium oxalate precipitates a crystalline double salt. Sodiumhyposulphite precipitates a boiling solution easily, but incompletely.Sodium acetate behaves similarly. The sulphates of potassium andsodium precipitate crystalline double salts, soluble in a saturatedsolution of the precipitant.The author describesin a previous paper (BUZZ. SOC. Chinz., 31, 486)the chloride, nitrate, and sulphate of scandium ; the double sulphates,the double oxalate, Sc,(C,O,),.K2C2O4.3H,O ; the acetate, the formate,and selenite, 3Sc,0,.10SeO2.4H2O.The existence of scandium was predicted by Mendelejeff, and its pro-perties described under the name of ekabor (Anmalen, Sq., 8, 1.13).The following table shows a comparison of the observed properties ofscandium witlh those predicted of ekabor.SC,(SO~),.~K$~O~, S C , ( S O ~ ) , .~ N ~ S O ~ . ~ ~ H ~ O , S C ~ ( SOi)s.(NH4)2SO4 ;8upposed Characters of Ekabor.Atomic Weight = 44.Ekabor should have but onestable oxide, Ebz03, :t strongerbase than alumina, which itshould resemble in mauy respects.It should be less basic than mag-nesia.Observed Ch,aracters of Scandium.Atomic Weight = 45.Scandium forms only one oxide,Sc203, more energetic than alu-mina, and less so than magnesiaINORGANIC CHEMISTRY.9Ekabor oxide should resembleyttria, although less basic. The se-paration of these two earthswill bedifficult, depending on differencesof solubility or of basicity.Oxide of ekabor is insoluble inalkalis; it will probably not de-compose ammonium salts.The salts should be colourless,and give gelatinous preci.pitateswith KHO, Na2C03, and HNaS02.The sulphate should form adouble salt with K2S04, havingthe composition of alum, but notisomorphous with it.But few ekabor salts shouldcrystallise well.The anhydrous chloride shouldbe decomposed by water, givingoff hydrochloric acid.The oxide is infusible a,nd so-luble with difficulty in acids afterignition.The density of the oxide wouldbe about 3.5.Scnndia is less basic than yttria,and their separation depends ondifferences of solubility betweentheir nitrates.Scandium hydrate is insolublein alkalis ; it does not decomposeammonium chloride.The salts are colourless, andgive gelatinous precipitates withKHO, Na2C03, and HNaS02.Potassium - scandium sulphateis anhydrous, but otherwise cor-responds in composition withalum.Scandium sulphate does notform distinct crystals, but thenitratc, acetate, and formatme crys-tallise well.The crystallised chloride is de-composed by heat, giving offhydrochloric acid.The oxide is an infusiblepowder, nearly insoluble in acidsafter ignition.The density of the oxide = 3.8.c. w. w.Absorption of Nitrogen Dioxide by Ferrous Salts. ByJ. GAY (Compt. rend., 89, 410).-Peligot assigned the formula4FeS04.N20, to the compound of nitrogen dioxide with ferrous su1-phate. The author finds that the composition of this body depends ont'he temperature and on the pressurc of the residual nitrogen dioxide.At temperatures up to 8" and at the ordinary pressure, the com-pound formed has the formula 3FeS01.N202 ; from 8" to 25", at theatmospheric pressure, i t has the formuh 4FeSO4.N2O2 ; at tempera-tures above 25" nitrogen dioxide is rapidly given off, and the com-pound 5FeSO4.N,O, is produced.All these compounds exhibit very marked tensions of dissociation,a fact which explains their decomposition in st vacuum ; they are alsodecomposed by a current of hydrogen.Reducing agents, such as ferrous oxide, reduce the nitrogen dioxide,a mixture of monoxide and free nitrogen being evolved, while thetemperature rises sensibly.c. w. w.Nitrosothioferrates. By J. 0. R l ~ ~ ~ (Ber. 12,1715-1717).-By the action of potassium nitrite and ammonium sulphide on aferrous salb, Roussin (Ann. Chim. Phys. [3], 52, 285) obtained a blacksubstance, which was afterwards examined by Porczinsky (Annden,125, 302), Demel (Be?-., 12, 46l), and Pawel (ihid., 12, 1407). Thisis named by the author ammoiziunz nitrosoferrclthiofen.ate10 ABSTRACTS OF CHEMICAL PAPERS.On the addition of an alkali ferrous oxide is precipitated, andpotassium nitrosothioferrate is obtained.The free acid which isliberated when this salt is treated with hydrochloric acid, combineswith alkaline sulphides to form a red salt, to which the name nitroso-ferrous potassimn sulphide is given.Salts correspondiug with each of the two first-mentioned acids havebeen prepared. They are all converted into nitroprussides by the actionof potassium cyanide. w. c. w.Potassium and Ammonium Ferric Chromates. By C.HENSGEN (Rer., 12,1656-1658) .-These salts separate out in dark-redplates containing 4 mols. H20 [K or NH,], when a solution containingferric chloride and ammonium or potassium dichromate, is slowlyevaporated. They have the formula KzCr04.Fe2(Cr04)3.4H,0. Theammonium salt is decomposed by cold water and also by the action ofheat.w. c. w.Contributions to the Chemistry of the Chrornammonium-compounds. By S. 31. JORGENSER (J. p r . Chem. [el, 20, 105-145.-1. C l ~ l o r o p u ~ p u ~ e o - c l ~ ~ o ~ ~ a i ~ ~ ~ ~ ~ Salts.-The starting point for thesesalts is the chloride, CI,(Cr210NH3) C1,. This is prepared by reducingviolet chromic chloride in a stream of pure dry hydrogen, at a redheat, and adding it to a solution of ammonium chloride in strongammonia (25 grams Cr,CI, reduced to Cr2C14, 90 grams NH4C1,0.5 litre ammonia). Air is then passed through the blue liquid untiloxidation is complete. Two litres of crude hydrochloric acid areadded, and the mixture is boiled for some minutes, during whichchloropurpureo-chromium chloride separates as a carmine-colouredpowder.The crude chloride is washed with a mixture of equal volumesof hydrochloric acid and water, dissolved in very weak sulphuricacid, and filtered into a great excess of strong cold hydrochloric acid.The resulting precipitate is boiled with hydrochloric acid, and washedfirst with a mixture of acid and water, then with alcohol, andfinally dried in the air a t the ordinary temperature. This chloride isalso a bye-product in preparing ClAve’s tetramine chloride by thefollowing process :-Ammonium dichromate is I educed by boiling withhydrochloric acid and alcohol, and after addition of ammoniumchloride the liquid is evaporated to dryness. The dry residue is thendissolved in strong ammonia ; strong hydrochloric acid is added, andthe crystals which are deposited on standing are washed first with amixture of equal parts of hydrochloric acid and water until free fromammonium chloride, then with water, and finally dried.It consistsof a mixture of chromium-tetramine chloride and chloropnrpureochloride. This mixture must be protected from the action of lightduring the remaining operations. It is dissolved in cold water, andshaken with a solution of one part of ammonium sulphate in five partsof water. The tetramine chlorosulphate precipitates in crystals ;the filtrate containing the purpureochloride is mixed with hydrosilico-fluoric acid, and gives a precipitate of chloropurpureo-chromium silico-fluoride. After being washed, it is treated with dilute hydrochloriIN ORGANIC CHEMISTRY.11acid, to reconvert it into chloride ; after reprecipitation with strongacid and washing, first with dilute acid and then with alcohol, it isquite pure. The two salts may also be separated by taking advantageof the insolubility of the componnd C1,( CrJONH,) ( HgsC1,),, producedby adding mercuric chloride to the mixture. The mercury-compoundatter washing is easily reconverted into the chloride by treatmentwith h-j-drochloric acid.Chloropurpureo-chromium chioride is a red crystalline powder, of apurer red colour than the corresponding cobalt-compound. It appearsto crystallise in octohedra of sp. gr. 1.687. It dissolves in 154 partsof water a t 16", and forms a violet-red solution, which, on exposure tolight, deposits chromium hydrate.When it is kept, even in the dark,or boiled, roseochromium chloride is produced. It gives the follow-ing reactions :-With sodium hypochlorite, nitrogen is evolved, andthe chromium is oxidised to chromic acid. Its solution gives a preci-pitate with strong hydrochloric acid, owing to the insolubility of thechloride in acid. With hydrobromic acid, it gives a crystalline preci-pitate of the bromide, and with solid potassium iodide one of theiodide. When boiled with potassium cyanide, it turns yellow. Strongnitric acid precipitates the chloro-nitrate. Hydrosilicofluoric acidthrows down the red crystalline chlorosilicofluoride. Platinic chlorideprecipitates, even from a very dilute solution, the chloropurpureo-chromium platinochloride.Sodium platino-bromide gives an ana-logous precipitate. Mercuric chloride gives red needles of the doublesalt. Precipitates are also produced by potassium mercuribromideand iodide, by sodium dithionate, potassium chromate, and dichromate,ammonium molybdate, and phosphomolybdate, and by picric and oxalicacids. I n these respects this salt closely resembles the analogouscobalt salt. On treatment with silver nitrate only four atoms ofchlorine are removed, and the chloro-nitrate is formed. By rubbingthe solid chloride with silver oxide and water, raseochromium hy-drate is formed. It is a, deep red alkaline liquid, which gives ayellowish-red precipitate of roseochromium bromide with stronghydrobromic acid ; this, when boiled with hydrobromic acid, changesto bromopurpureo-chromium bromide.In the chloro-chloride, theradicle chlorine is so firmly combined that hot strong sulphuric aciddoes not expel i t ; the product is acid chloro-sulphate.Towards reducing agents, however, the chromium series differ inbehavionr from the cobalt series, for the chromium is not so easily re-duced. With snlphuretted hydrogen, or with ammonium sulphide,the purpureo-cobalt-compounds give cobalt sulphide, but the purpui-eo-chromium compounds suffer no change, except the forniation of acrystalline pnrpureopolysulphide, if the ammonium sidphide containsmuch free sulphur. The cobalt salts are also reduced by potassiumferrocyanide, whereas the chromium salts give a precipitate of ferro-cyanide of chloropurpureo-chromium.The latter part of this paper is occupied with detailed descriptionsand analyses of numerous salts of chloropurpureo-chromium chloride,prepared by double decomposition.They have all a red or orange-redcolour, and closely resemble the corresponding salts of chloropurpureo-cobalt. W. R12 ABSTRACTS OF CHEMICAL PAPERS.Behaviour of Copper- Ammonium Chloride with FerrousSulphide. By W. F. K. STOCK (C'hern. News, 40, 159).--In thecourse of recent experiments on the accurate determination of carbonin iron and steel containing much sulphur, i t appeared desirable toascertain definitely in what manncr the reagent used for the carbonseparation acted on the iron sulphnr compound, but as the compositionof that compound is unknown, it was thought best t80 experiment witha sulphide of known quality.The process used for the carbon separa-tion was McCreath's method based on treating a weighed quantity ofiron or steel with a hot concentrated solution of copper ammoniumchloride.From the results, it was evident that the actions of the double cop-per ammonium salt upon iron carbide and upon iron sulphide were pre-cisely analogous, and that tbe method held out no hope of separation.It only remained to find to what extent the decomposition had pro-ceeded during the exposure, which was for half an hour a t nearlyboiling heat, It is shown that allowing for oxidation during washing,&c., it may safely be assumed that 80 per cent. of the original sul-phide was decomposed by the double copper salt with liberation of thecorresponding amount of free sulphur.A second experiment was made with native ferric sulphide, whichwas very finelypowdered and exposed a t boiling heat for over an hourt20 the copper solution, but although some free sulphur was obtained,the deconiposition was far from complete.Action of the HaloTd Acids on the Sulphates of Mercury.By A.DITTE (Ann,. Qhim. Phys. [ 5 ] , l7,120--128).-1t has beenstatedthat dry hydrochloiic acid gas decomposes mercuric sulphate, formingmercuric chloride and free sulphurie acid, and t h a t since the chlorideis more volatile than the acid, the former can be separated by sublima-tion at a suitable temperature ; and further, that hydriodic and hydro-cyanic acids act in a similar manner. The author shows that thesestatements are wholly incorrect.When dry hydrochloric acid is passed over mercuric sulphnte a tordinary temperatures, no reaction ensues : on warming the sulphate,absorption takes place, with disengagement of heat and without forma-tion of water ; on heating more strongly, the product sublimes, butthe crystals hare no resemblance whatever to sublimed mercuric chlo-ride. An analysis of the crjstals showed that their compositionexactly corresponded with the formula HgSO, 2HCl; they are veryhygrometric, dissolving in water apparently without decomposition.When volatilised they do not disengage hydrochloric acid.Hydrobromic acid gas acts in a precisely similar manner, formingthe compound HgS01.2HBr.The body HgSO,.BHCl is likewise formed with great facility bygently heating a, mixture of sulphuric acid and mercuric chloride, inmolecular proportions ; or by dissolving the neutral sulphate in hydro-chloric acid, and evaporating until crystals are obtained.The action of hydriodic acid is different; sulphuric acid decom-poses mercuric iodide on heatir,g, no compound o f thc formula€€gSOa.2HI being formed. In the same manner, solution of hydriodicD. BMINERALOGICAL CHEMISTRY. 13acid in excess, partly or wholly decomposes mercuric sulphate, but nodefinite combination takes place.Hydrofluoric and hydrocyanic acids are without action on mercuricsulphate.Basic mercuric sulphate, twpeth wahzeral, acts with regard to hydro-chloric acid in a manner analogous to mercuric sulphate, but it ab-sorbs 6 molecules of HCl for every molecule of sulphate, forming thecompound HgS04.2Hg0.6HC1 ; the latter on being heated strongly,breaks up into the mercuric compound and mercuric chloride,Hg3S06.6HC1 = HgS04.2HCl + 2HgC12 + ‘2H20.thus :.-A precisely similar compound is formed by the action of eithergaseous or liquid hydrobromic acid on tnrpeth mineral.A New Salt of an Iridammonium. By K. BIRNBAUV (Bey.,12, 1544--1547).-By boiling the double salt of iridic sulphite andsodium sulphite with hydrochloric acid, an acid salt is formed pre-sumably of the formula Irz( S03)3.3NaHS03. When its solution issaturated with gaseous ammonia, a compound crystallises out in redcrusts, having the formula Ir2Naa( S03)G(NH3),.10H20. The authorassigns to it the constitutional formula-SO3 : Ir(NH3)3 NH4Naso,< 1SO3 : Ir(N%)sJ. W.+ 3 >s03.0H2c),and supposes the SO3 group t o be i n combination with an irid-ammonium of the formula (NH&Ir2. W. R