As a further confirmation, the percentage of nitrogen was determined volumetrically, and found to be 80.48 cubic inches per 100 grs. of blue nitrate. Reduced to weight it gives— The green hydrochlorate appears under the microscope as feathery stars, not so readily soluble in water as the blue nitrate. The formula of the blue nitrate at once suggested that of the green hydrochlorate, namely : 6 2N2H PtC = 412; It was a far more difficult task to determine the composition of the coppery precipitate given by Raewsky's nitrate with an acid solution of protochloride of platinum. Qualitative experiments tell much with regard to it. The following facts are readily ascertained: 1. Boiled in water, it is resolved into Magnus' green, Gros' nitrate, and protochloride of platinum-the two latter in equal number of atoms. 2. The proportion of platinum to chlorine is one atom of each, just as in the green compound of Magnus. 3. If a reducing agent such as sulphate of iron be added to the solution, instead of boiling the liquid, the quantity of Magnus' green is sensibly increased, owing to the reduction of Gros' nitrate to hydrochlorate of diplatosamine, which, with the protochloride of platinum present, occasions the additional amount or the green precipitate. 6 4. These reactions plainly point to a formula of this kind x(N2H Pt2Cl2)O.NO. The difficulty has been to find the value of x, and this has been due to the excessive uncertainty of the composition of the various samples of the coppery body, as is shown at once by ascertaining the percentage of platinum, which varies from 62 in the worst to 56 in the best looking product. Then, again, it was found difficult to bring the salt into solution unchanged in composition for graduation by permanganate. This appears to be best effected by digesting the compound in a solution of Raewsky's nitrate containing excess of dilute sulphuric acid. Determinations of the proportion of platinum in the platinic state have given numbers varying from 1 in 6 of total platinum in the best, to 1 in 10 in the worst. From these uncertain results I am only able to give a theoretical formula, being prevented from making any further experiments for some time to come; but of the correctness of the formula I feel pretty confident. Theoretical composition of the coppery salt it is thus seen to have a similar composition to that of the platinid5(KCy.PtCy) cyanides* {(KCy Pty.c1}, & of the platinum in each being in the platinous, and of each in the platinic state. The decomposition by boiling in water would be 2(N2H PtO.NO5) + 2PtCl2 + 4N2H¿Pt2Cl2 All the experiments alluded to were made with the coppery salt, prepared by mixing Raew sky's nitrate with a solution of protochloride of platinum highly acidified with nitric acid; a far better method, more easy, more profitable and more sure, has only lately been discovered. This consists in immersing the green compound of Magnus for some hours in a highly dilute solution of Raewsky's nitrate strongly acidified with nitric acid, and in large excess; the green salt then expands greatly in bulk and becomes brilliantly coppery. A brilliant sample of this salt gave 59.2 per cent. of platinum; the above formula requires 60.9. On boiling the product in water, the green compound of Magnus which was formed contained platinum equal to 40 per cent. of the weight of the salt, thus corresponding exactly with the decomposition which occurs when the coppercompound is simply boiled with water, in which case of the platinum separates as the green salt of Magnus. In the formation of this coppery salt, by the above method, * Chem. Soc. J. vol. xiii, 1861, p. 113. a rather singular decomposition of Raewsky's nitrate occurs thus NH12Pt2CIO3.2NO, — 0 = N2H PtO.NO + N2H PtCIO.NO, 12 6 5 6 or by the loss of an atom of oxygen, which it imparts to the Magnus' green, it is reduced to equal numbers of atoms of Gros' nitrate and nitrate of diplatosamine. Sulphate of soda will show the presence of the former, and nitrous acid that of the latter salt. The same reaction can be shown in another way. Raewsky's nitrate and sulphate of soda may be left mixed without change; but on adding a small quantity of a reducing agent, such as protosulphate of iron, the characteristic crystals of Gros' sulphate appear in a few minutes. The singular composition of these coppery bodies seems to point towards the existence of an oxide of platinum having the composition PtO,, I have not as yet obtained any proof of its existence, but I cannot help thinking it will one day be indicated by some characteristic of colour or form, &c., distinguishing it from the protoxide and peroxide. There are one or two isolated facts which I may as well insert here by way of conclusion. On one occasion wishing to prepare the green salt of Magnus, protochloride of platinum was prepared with sulphurous acid as usual, the solution was divided into equal parts, and one-half was converted into hydrochlorate of diplatosamine by ammonia and mixed with the other half, when, instead of the usual green, a dove-coloured precipitate fell; it was not altered by boiling, simply dissolving and crystallising out of the same colour. On determining the platinum, it corresponded perfectly with that in the green compound— Green compound gave Pt = 64.3 Dove-coloured compound, Pt = 64.62. On boiling in hydrochloric acid, however, it became quite green, and on testing with zinc and hydrochloric acid, sulphuretted hydrogen came off directly, proving the presence of sulphurous acid. To determine the amount of sulphurous acid, the dove-coloured salt was dissolved in ammonia and graduated by permanganate. It was found that for every 2 atoms of platinum, 2 atoms of oxygen were absorbed, which points to a formula for the doveN2H&PtC PtCl N2H6PtCl PtO.SO2 coloured salt = 2 I must also here remark that since writing this paper, I have looked into Graham's Chemistry on nitrous acid, and I was not a little gratified to find that he ascribes to nitrous acid a thoroughly basic character, a view I had not noticed in his or other works. XXXIV.-On a Cyanogen-derivative of Marsh-gas. By HENRY BASSETT. THE action of cyanide of potassium on chloropicrin appeared, from the other reactions of the latter substance, to promise interesting results. The experiments made in this direction have led to the production of a substance representing chloropicrin in which 2 atoms of chlorine are replaced by cyanogen. The process adopted was as follows:- 15 grms. of the best fused cyanide of potassium are dissolved in 1 oz. water, oz. alcohol added, and then 10 grms. chloropicrin. The flask containing the mixture is fitted with a long condensing tube, and then placed on a water-bath till the reaction is started. It must then be quickly removed, and the upper end of the tube inserted into an opening in the chimney, to avoid inconvenience from the pungent vapours. The liquid rapidly becomes dark red, and a very violent reaction takes place for about ten minutes, sometimes necessitating the cooling of the flask towards the end of that time. The quantities mentioned cannot be safely exceeded, and the proportion of alcohol must not be sufficient to dissolve the chloropicrin, as the action then becomes almost explosive. The contents of a sufficient number of flasks, when cold, are agitated with ether, and the resulting dark red ethereal solution is separated. The watery liquid remaining is dark brown, from the presence of a considerable quantity of the substances known as azulmic or paracyanic acids, and contains chloride, nitrite, carbonate, and oxalate of potassium, and large quantities of ammonia. The ethereal solution is heated in a retort in the water-bath, to separate the ether and alcohol as far as possible, and the heat should not be continued too long, the substance being thereby altered. The residue is mixed with water and filtered from some insoluble matter, the filtrate saturated with chloride of sodium, agitated with ether, separated, and the ethereal solution evaporated in an open vessel at very gentle heat. The substance thus obtained is semi-fluid, of a dark red colour, does not solidify at - 20°, has a peculiar and unpleasant smell, and is soluble in all proportions in water, alcohol, ether, chloroform, &c., the solutions staining the skiu bright yellow. It retains a considerable quantity of water, which is not given up over sulphuric acid in vacuo; neither can it be dried by heat, as even at ordinary temperatures it slowly alters, and near 100° it is completely decomposed. In the solution obtained by boiling it with potash, the presence of chloride, nitrite, and cyanide is readily detected by the usual tests, and large quantities of ammonia are formed. It was obviously useless to attempt the analysis of this substance, but I think its composition is sufficiently indicated by the analyses of its compounds with oxide of lead and nitrate of silver. The watery solution was mixed with excess of acetate of lead, filtered from any slight precipitate, and ammonia cautiously added, taking care not to use sufficient to render the liquid alkaline; the bright orange precipitate washed with boiling water and dried at 100°, bears a high temperature without decomposition, and is extremely hygroscopic. I. 32 grm. gave 3615 grm. Pb SO. II. 1.01 grm. gave '18 grm. AgCl (dissolved in cold dilute nitric acid, nitrate of silver added, and then heated with excess of strong nitric acid till completely decomposed). III. 3.08 grms. gave 509 grm. CO2 and '07 grm. H2O. Pb"SO4. grm. (another preparation) gave 3585 grm. V. 994 grm. gave 184 grm. AgCl. VI. 1·113 grm. gave 28.3 c. c. N. at 0° and 760 mm. These results lead to the formula CNO2CICy2+ 3Pb"O. |