26 ABSTRACTS OF CHEMICAL PAPERS INORGANIC ANALYSIS. Estimation of Calcium in Ash of Forage Plants and Animal Carcases. S. B. Kuzirian. ( J . Amer. Chem. SOC., 1916, 38, 1996-2000.)-Phosphorus is re- moved from the acid solution of the ash as ammonium phosphomolybdate. Calcium is precipitated as oxalate, t-he solution containing the precipitate is boiled for half an hour, the oxalate filtered off on a Gooch crucible, and ignited to render molyb- denum compounds insoluble. The ignited precipitate is treated with dilute hych- chloric acid, the solution filtered, iron and alumina separated as hydroxides, and the calcium in the filtrate finally precipitated as oxalate.0. C. J. mciency of Calcium Chloride, Sodium Hydroxide, and Potassium Hydroxide as Drying Agents. G.P. Baxter and H. W. Starkweather. ( J . Amer. Ckm. Sm., 1916, 38, 2038-2041.)-Experiments are described which show that potassium hydroxide a t 25" C. is as efficient a, desiccating agent as sulphuric acid, the vapour pressure of the lowest hydrate being about 0.002 mm. That of sodium hydroxide is 0.15 mm., and of calcium chloride 0-34 mm. a t 25" C. The temperature co- efficients are high, the vapour pressure of sodium hydroxide being 0.04 mm.at 0" and 1.15 rnm. at 50°, and that of calcium chloride 0.07 mm. at 0" and 1-34 mm at 50" That of potassium hydroxide at 0" was unmeasurable by the methods used; at 50" it is 0.007 mm. G. C. J. Estimation of Hydrogen Sulphide in Water. L. W. Winkler. (Zeitsch. anqew. Chem., 1916, 29, 383-384.)-The water is collected in a 500 C.C.stoppered bottle containing about 20 grms. of marble free from any trace of sulphide, and the bottle should be filled completely with the water. The stopper of the bottle is removed and replaced by a rubber stopper carrying the stem of a bulb tube which is provided with a side tube; a tapped funnel containing hydrochloric acid is attached to the top of the bulb tube, and the side tube is connected with a U-tube contahing cotton wool, and then with an absorption vessel containing acid-free bromine water.The hydrochloric acid is admitted gradually into the bottle, and the liberated carbon dioxide passes over into. the absorption vessel, carrying with it the hydrogen sul- phide conbined in the water. The contents of the absorption vessel are then evaporated until all free bromine and hydrobromic acid have been expelled, and the residual solution is titrated with & borax solution, using methyl orangeINORGANIC AKALYSTS 27 as the indicator.The oxidation of the hydrogen sulphide proceeds according t o the equation: H,S+4Brs+4H,0=H,S0,+8HBr, and each C.C. of & borax solution is equivalent to 04705 mgrm. of hydrogen sulphide.Alternatively, the free bromine alone m y be expelled by boilirg, and the hydrobromic and sulphuric acids then titrated with & borax solution; in this case 1 C.C. of & borax solution (19.11 grms. of the crystallised salt per litre) corresponds to 0-341 mgrm. of hydrogen sulphide. Hydrogen sulphide in relatively strong aqueous solution may be estimated as follows: A stoppered flask containing 100 C.C. of & permanganate solution and 1 grm.of pure sodium hydroxide is weighed; a few C.C. of the hydrogen sulphide solution are now introduced, and the flask is again weighed. After one hour the mixture in the flask Is acidified with aulphuric acid, potassium iodide is added, and the liberated iodine titrated with & thiosulphate solution.Each C.C. of & permanganate soh- tion is equivalent to 0.4263 mgrm. of hydrogen sulphide. w. P. s. Method of Extraction as affecting the Estimation of Phosphoric Acid in Soils. H. Hale and W. L. Hartley. ( J . Id. and Eng. Chem., 1916,8,1028-1029.) -Experiments are recorded which show that two hours’ digestion with 2N nitric acid extracts as much phosphoric acid as ten hours’ digestion with hydrochloric acid of sp.gr. 1.115. It is also shown that the ten hours’ digestion with hydro- chloric acid brings down m.uch more interfering substances than the shorter digestion with nitric acid. G. C. J. Reagents for Use in Gas Analysis. V.-Relative Advantages of Sodium and Potassium Hydroxides in Preparation of Alkaline Pyrogallol. R. P. Anderson. ( J .In& and Eng. Chem., 1916,8, 999-1001.)-A criticism of the con- clusions of Shipley (AXALYST, 1916, 41, 349), who advocated the use of sodium hydroxide. It is agreed that when the best practicable solutions of sodium and potassium pyrogallate are eompared, the specific absorption of the sodium reagent is the higher, and the cost of the sod-urn reagent is less, even in normal times.But the author does not agree that the use of sodium hydroxide effects any real economy, even a t present prices of potassium compounds, as the loss of time when using the sodium reagent much more than counterbalances its lower cost. He gives figures showing that, a saving of three shillings in cost of chemicals involves the expenditure of some thirty addj tional hours in manipulation, each analysis with sodium pyrogallate taking about two and a half minutes longer than would be the case were potassium hydroxide substituted for sodium hydroxide.G . C. J. Solution of Red Lead. G. Torossian. ( J . Id. and Eng. Chem., 1916, 8, 1076.)-Dilute (1 : 5) nitric acid containing 0.5 per cent. of tartaric acid is an excellent solvent for red lead. Th.e hot mixture dissolves red lead instantly.Stronger soh- tions are sometimes convenient, but do not keep well. G . C. J.28 ABSTRACTS OF CHEMJCAL PAPERS Volumetric Estimation of Tin. R. L. Hallett. (J. Soc. Chem. Id., 1916, 35, 1087-1089.)-The following modification of the method of Pearce and LOW (Low, Technid Metho&s of Ore Analysis, p. 208) gives results accurate to within 0.1 per cent.in about one and a, half hours. It is based on the oxidation of stannous chloride to stannic chIoride by means of iodine in the presence of cold hydrochloric acid. Sulphuric acid in small amount does not interfere with the titration, but nitric acid or nitrates must not be present When tin is dissolved in hydrochloric acid it will require complete reduction to the stannous condition before titration. The final solution should contain between 25 and 40 per cent.of strong hydrochloric wid, and during the reduction and titration a current of carbon dioxide should be passed through the flask. From 0.5 to 2 grms. of the sample are dissolved in about 50 C.C. of hydrochloric acid, and the solution is diluted to 200 C.C. with water, and gently boiled for thirty minutes in a flan9k in which is suspended a nickel coil made by rolling 6 sq.inches of heavy sheet nickel into a loose roll, while the mouth of the flask is covered with a watch-glass. The reduc$ion is indicated by the change in the colour of the liquid from yellow to pale green, and the Aask is then left to cool, after the introduction of two small cubes of crystalline marble to produce an atmosphere of carbon dioxide.The nickel coil is withdrawn and washed with dilute hydrochloric acid (1 : 3) during its withdrawal, and the solution is a t once titrated with standard iodine solution, with starch as indicator. A convenient strength for the iodine solution is 10.7 grms. in 1,OOO C.C. of water containing 20 grms. of potassium iodide (1 c.c.=I per cent.of tin). Nickel, cobalt, manganese, molybdenum, uranium, chromium, aluminium, zinc, lead, calcium, magnesium, sulphates, phosphates, bromides, iodides, and fluorides, have no effect on the results, unless present in sufficient quantity to mask the colour of the indicator. Arsenious and antimonious acids have also no influence in the presence of hydrochloric acid of the strength used in this estimation.Precipitation of metallic antimony by the nickel coil may be prevented by the addition of an extra 25 C.C. of strong hydrochloric acid. Traces of copper do not influence the results, but if more than 0.05 grm. be present it should be removed beforehand by treatment with nitric acid, as in the case of titanium (infra). Bismuth is precipitated in metaIlic form in the reduction process, but the precipitate does not materially affect .the titration.Tungsten is also precipi- tated as a, bIue oxide. In small quantities the effect of this is negligible, but, if necessary, the solution may be filtered, and again reduced before the titration. Titanium is best removed by converting the tin into oxide by evaporation with nitric acid, fusing the residue for five minutes with potwsium acid sulphate, dis- solving the melt in dilute sulphuric acid, and filtering the solution.Any tungsten present may then be separated from the tin by treatment with ammonium car- bonate solution, in which the tin is insoluble. Ferrous chloride is not oxidised during the titration, unless a large amount of iodine is added and allowed to stand for some time.C. A. M.APPARATUS, ETC. 29 APPARATUS, ETC. Melting-Point of Fats. (Giorn. di Parrn. e di Chirn., 1916,4,151-153; through Chem. Eng. and Manufac., 1916, H, 188.)-To the thermometer employed is attached a platinum wire of a thickness of 0.3 to 0.4 mm., the free end of which is formed into a loop of 8 to 9 mm. diameter. The wire is attached in such a manner that the loop assumes a, vertical position immediately in front of, but not touching, the mercury bulb. The fat to be examined is melted at a; gentle heat, and when it is almost cool enough to solidify, the platinum loop is dipped parallel to the surface into the liquid and quickly withdrawn, leaving a thin film of fat, completely filling the loop. This is allowed to cool for an hour, the wire attached to the thermometer, which is:then immersed in a wide-necked glass flask of about 250 mm. capacity filled with distilled water, which is slowly heated on an asbestos card, until the film of fat in the platinum loop becomes completely transparent just before breaking up. The temperature of complete transparency is the melting-point of the fat. H. I?. E. H.