A n a l y t i c a 1 C h e m i s t r y. Iodometric Estimation of Telluric acid. By F. A. GOOCH and J. HO~T~LAND (Zeit. aizorg. Chern., 1894, 7, 132-136).-l1elluric acid is reduced by hydriodic acid, and iodine is set free ; but when the free iodine is titrated with sodium thiosulphate, from 20-26 per cent., more iodine is found than that required for the reduction to tellurous acid. Telluric acid may be accurate1.y estimated as follotvs. The telliiric acid is treated with rpotassium bromide and dilute sulpliuric acid in n distillation apparatus, and a stream of carbonic anhydride passed through tlie boiling solution in order to remove the bromine. The latter is conducled into a solution of potassium iodide and the liquid titrated with sodium thiosulphate. The resnlts agree very well with one another and with the acceptaucc of 127 as tlie atomic weight of tellurium.The authors prepare their standard solution of tellurium as follows. Telliirium dioxide, obtained by dissolving pure tellurinm in nitric acid and lieat,irig the product at a red heat, is dissolved in strong potassium hydroxide and precipitated with dilute sulphuric acid. The precipi- tate is dissolved in sulphuric acid (diluted one half), treated with excess of potassium permatiganate, neiitralised with oxalic acid, and the excess of oxalic acid carefully destroyed with dilute permanga- nate solntion. A series of twelve determinations of the equivalent weight of tellurinm dioxide by Brauner’s method (Trans., 1891, 258), gave the molecular weight of telliirium dioxide = 159, and the atomic weight of tellurium = 127 (0 = 1 G ) .E. C. R. Volumetric Estimation of Phosphorus in Steel. By W. A. NOYES and E. L>. FROHMAN ( J . Amw. Chem Soc., 1894, 16, 559-559). --The method proposed by Dudley and Pease (dbstr., 1893, ii, 553) is rapid and easy of execution and gives concordant results. There is, however, an uncertainty of about 10 per cent. as regards the factor which should be used for converting the iron equivalent of the per- manganate solution into the phosphorus equivalent. The author’s experiments conclusively prove that if the reduction of the molybdate compound by means of zinc and sulphuric acid is properly executed the metal exists in a form corresponding with the oxide Mo,O,. The conditions are (i) that the passage of the solution through the reducing flask should be preceded by that of some dilute acid, so t,hat most of the air in the reduciag flask may be expclled before the solu- tion enters i t ; (ii) that the diluted solution shall pass through the reduciiig flask slowly and be followed by the acid wash-water with- ont sllowirig any air t o enter a t the top ; and (iii) that the reduced solution shall be titrated at once and without further dilution.It should be of a pure green, cot an olive-green colour. It is now a thoroughly established fact that in the yellow animo- nium phospho-molgbdate t!iere exist twelve molecules of molybdic anhydride t o one atom of phosphorus. The quantity of the latter mayANALYTICAL CHEXISTRY. 31 therefore be ~ e i = y correctly cslculat,ed from the amount of per- niangnnate required to reosidise the Mo,O, to ?Moo3.L. DE K. Analysis of an Apatite in a Laminated Graphite from Ceylon. By P. JANUSCH and J. Locm ( Z e i f . nnorg. Chern., 1894, 7,154-157). -The analysis was conductcd as follows :-A4bout 0.5 gram of the mineral is dissolved in stroxg nitric acid ancl evaporated t o dryness with mercuric oxide. The residue is moistened with nitric acid, dis- solved in water, heated to boiling, and neutralised with ammonia, whereby the phosphoric acid is precipit:i tcd as mercuric and calcium phosphates. The precipitate is washed, heated over a bare flnme, dissolved in dilute nitric acid, aud precipitn t,ed with molybdenum solution. The filtrate is concentrated, the excess of molybdic acid filtered off and the remainder of the molybdenum pi-ecipitatcd as sul- phiclc ; the iron ancl aluminium aye theil estiniated iii the solution.The filtrnts from the iron and aluminium contains manganese, magnesium nncl calcium ; these are separated by a method previously described (Zcit. nuorg. Cl~cnz., 6 , 57). The fiitratc from the mercuric and csl- cium phosphates contains a, small q umti ty of calcium and the alkalis. It is evaporated to dryness, heated until the mercuric oxide is sepa- rated, and then dissolx-ed in liydrochloric acid ; the calcium is deter- mined as oxalate, the potassium by platinic chloride, and the sodium weighed as sulphate. The chlorine in the apatite is dctemiinecl b y dissolving a portion i n nitric acid containing silver nitrate : the fluorine by heating with con- c e n h t e d sdphuric acid in n platinum apparatus and collecting the hydrogen fluoride evolved in pure sodium hyclrosicle ancl precipitating as calcium fluoride.E. C. R. Detection and Estimation of Small Quantities of Arsenic in Copper. By F. A. GOOCH and H. P. Mcistcrx\- (Zeit. nnorg. Chenz., 1%)4, 7, 127--131).--T11e estimation is bnsed on a inetbocl described by Sanger (Ahstr., 1892, 382), whereby the mirror of arsenic ol3tained is compared with n, series of standard mirrors. Not more than 1 gram of the copper is diysolved in dilute nitric acid, 2-3 C.C. of strong sulpliuric acid is added and the solution evaporated until fumes of sulphuric acid are et.olved. The concentrated residue is then dis- tilled with strong, hydrochloric acid and potassium bromide, and the distillate collected in dilute hyclrochloric acid.The distillate is again distilled with a fresh quantity G f potassium bromide, and the final distillate is treated with n small quantity of stannous chloride dissolved in hydrocliloric acid in order to convert m y free bromine into hydro- bromic acid ; it is now ready for the Marsh’s apparatus. The resu1t.s obtained by the authors shorn that, 0.05 milligram of arsenic may be determined by this process mitli fair accuracy ; and they gire the results of analyses of commercial copper contailling from O.Ol-O*OOO5 Apparatus for the Estimation of Carbonic Anhydride. By A. C. CHRISTOJIBKOS (Ber., 1894, 27, 2748-2751) .-The apparatus de- scribed by the author is a modification of Bnnsen’s i.vell-knon.n appa- per cent.of arsenic. E. c. R.32 ABSTRACTS OF OHEMCOAL PAPERS. mtus, sclphnric acid being substituted for calcium chloride as the drying agent, and a more stdble form given to the whole. A. H. Estimation of Carbonic Anhydride in the Air. By F. KRAT. SCHHER and E. WIENER (Il-iorintsh., 1494.15, 429432).-The method described in this paper depends on the fact that when (L solution of aodium carbonate is titrated with sulphiiric acid at the ordintmy temperature, using phenol ph tha lejin a s indicator, decdorissltion of the latter takes place as Roon as one-half of the sodium carbonate Ilns been decomposed with formation of sodium sulphate aud sodium hydrogen varbon*it-. With a mixture of solutions of sodium hydrox- icie anti carbonate, decolorisation takes place as scon a9 the whole of the former arid half of the latter have been neutrslised.If, there- fore, equal voluriies of a solution of sodium hydroxide be titrated with sulphuric acid, the one initncdit~tellp, aud the other after shaking with a, known volume of air cont~aining carboriic anhydride, the differ- ence i n the a.rnoiinh of acid require11 i n the two cases is equivalent to cme-hHlf of the amount of podium carbonate formed by the carbonic anligdride. The aiithors employ ~t solution of sulphuric acid of such strength that 1 C.C. is equivalent to 1 milli ram of carbonic anhydride, and the niimhei* of rntlligrrtms of carbonic anhyclride in any volume of air I;tiken is found by doubling ttie number of cubic centimetres re- presenting the difference Let ween the two titrations.H. G. C. Indirect Methods of Analysis. By H. SCHIFF (Qazzettu, 1894, 24, ii, 30--:16).--In cotinwtion with the work of E’orte (Abstr., 1894, ii, 352), the aathor refers to his own papers (Annulen, 105, 219 ; 107, 69) on indirect met Iiods of analyais. Avoioance of Certain Causes of Error in Hydrotometric Analysis. By A BO~OLETTI (C=’u:zetfu. M94,24, ii, 26--30).--0n ap- plying Clrti.k’s soap test to very hnrd waters, running. the soap solution into the uudiliittvl sattiple, rtritl also iiito diluted portions, the results obtained art, not concord:btit among thewselves. It’ the water is rich in earthy matter, it should be diluted wit,h distil1c.d water until the hiLr(intb8Y is rrbout, 22’ (parts of calvium car4)otintt: per 100,000 parts of water). If the water 18 soft. the rcsuhs obtained by the soap test make it seem too hard. In this case, the staridarsd mdution of calcium chloride, whose hardness is kiiown, is diluted uutil of about the same degree of hai-dntw as tlint of t.he s:unple unfler examination; the hardness of this is then detet*mineci and is found to be greater than the extent of tlbe dilution indicates. Taking the degree of dilution of the stiindard int,o consitieratioll, the true tiardnt*ss of the water to be exaillined can be c.ilculrtted (cotnpare Neugebauer, Abstr., 1891, 116). W. J. P. W. J. P.