ANALYTICAL CHEMISTRY.IN presenting this Report of the year’s work in analyticalchemistry, the writer wishes to emphasise the fact that, from thenature of the subject, it is scarcely possible to fulfil the objectof the ‘I Reports ” as set forth in the “ Introduction ” to Vol. I.(1904). I n the first place, the text must necessarily be somewhatdisjointed, since, in the particular branch of chemistry underreview, it is much more difficult than in others to construct aconnected narrative, for prominence can seldom be given to anyparticular researches which tend to advance our knowledge of thesubject as a whole. The broad reasons for this are not far toseek. Although there are some notable exceptions, it cannot bedenied that the number of papers found in journals devoted toanalysis in which new chemical principles are brought forward isbut few, the greater part of the literary matter which has to beperused in writing a review on analytical chemistry dealing withwhat, may a t best be described as improvements rather thanoriginal discoveries.Moreover, among this majority are to be foundsome communications which are not only valueless but actuallymisleading. For these reasons, therefore, as well as from the factthat this report makes no attempt a t being exhaustive, it is obviousthat some selection must be made, and-apart from his ownexperience-the writer has based his selection on the internalevidence contained in each paper. I n so voluminous and widelydisseminated a literature, it is possible, indeed probable, thatseveral deserving contributions to the subject may not havereceived notice, either from inadvertence or errors in judgment,but it is hoped that these omissions may amount to a minimum,and that a judicious selection has been made.I n connexion with research in analytical chemistry, it may bementioned that an important step has been taken by the (( Societyof Public Analysts and Other Analytical Chemists ” in inauguratinga scheme for the investigation of analytical processes and ofproblems in analytical chemistry, and also for the revision of pub-lished methods and their extended study when desirable.UndeANALYTICAL CHEMISTRY. 181this scheme it is suggested that the investigations might be under-taken by some of the senior students working in the larger collegesand teaching institutions, and a fund has been established by theabove Society for defraying incumbent expenses.The only con-ditions are that accounts of the researches are to be published(subject to the approval of the Editorial Committee) in TheAnalyst, and, when possible, to be brought before the Society inthe form of communications.1Inorganic Ch emist ly.Qualitative.-The system of qualitative analysis, the first twopapers on which were published last year by A. A. Noyes andW. C. Bray: has been further developed by A. A. Noyes, W. C.Bray, and E. B. Spear.3 This third communication deals with themetals of the ammonium sulphide group. It is impossible toreview such a paper as a whole, covering as it does someeighty pages, but among other things the authors show thatwhen the precipitate obtained with ammonia and ammoniumsulphide is dissolved in acid, and the solution boiled with a mixtureof sodium peroxide, hydroxide, and carbonate, the members of theiron group, including nickel, cobalt,, thallium, manganese, titanium,and zirconium, are precipitated, whilst those of the aluminiumgroup, including glucinuni, zinc chromium, uranium, and vanadium,remain in solution as sodium salts.A somewhat similar scheme,involving the use of hydrogen peroxide and sodium hydroxide, issuggest.ed by E. Ebler.4 H. Caron and D. Raquet 5 suggest sodiumperoxide as a reagent for the separation of the elements of group 111in presence of phosphates.The same compound is recommended byD. F. Calhane for the detection of chromium in presence of iron;the oxidation proceeds to the perchromate stage. For the rapidqualitative analysis of a mineral, W. B. Pollard’s 7 process consistsin fusing it with vaselin and sodium peroxide. The more posi-tive elements remain in the melt as insoluble oxides and carbonates,whilst the others exist in the highest state of oxidation as solublesodium salts; special tests must be made for mercury (on account ofits volatility) and for sodium. J. H. Walton, jun., and H. A.Scholz8 describe a method for decomposing certain slags andAnal@, 1908, 33, 41.Bid., 1908, 30, 481 ; A., ii, 538.Zeitsch. anal. Chen?., 1908, 4’7, 665 ; A . , ii, 987.BUZZ. SOC. chinz., 1908, [iv], 3, 622 ; A ., ii, 630.J. Amer. Chem SOL, 1908, 30, 770 ; A., ii, 630.Chem. News, 1908, 98, 211 ; A . , ii, 1069.Amer. Chem. J., 1908, 39, 771 ; A., ii, 732.’ J. Aster. Chem. SOC., 1907, 29, 137 : A . , 1907, ii, 391182 ANNUAL REPOKTS ON THE PBOGRESS OF CHEMISTRY.sulphide ores, which consists in fusing them with a mixture ofsodium peroxide, zinc sulphide, and potassium persulphate.In order to detect helium in minerals, F. BordasQ heats thesubstance, and passes the gas through a Dewar’s charcoal tube;the helium is much less readily absorbed @an other gases, and maybe detected spectroscopically.E. Selvaticilo advocates the use of thioacetic acid instead ofhydrogen sulphide for the precipitation of the metals of group 11.He also gives a scheme of separation obviating the use of ammoniumsulphide.H. Bollenbach 11 makes use of ammonium persulphatein separating the metals of group 11. G. D. Lander and H. W.Winter12 deal with the detection of poisonous metals.L. Tschugaeff 13 has pointed out that Pozzi-Escot’s molybdatemethod of detecting nickel 14 is less sensitive than his own dimethyl-glyoxime method.15 Pozzi-Escot 16 has modified his molybdatemethod, and more recently 17 he has described conditions wherebycobalt may be detected in the presence of 1000 times its amountof nickel. H. Grossmann and B. Schuck18 state that this methodhad been previously described by Marckwald; they also state thatas a test for nickel in the presence of cobalt it is less delicate thantheir O W ~ , ~ Q besides which there is a tendency for the precipitationof cobalt as the violet molybdate.20 H.Grossmann and W.Heilborn 21 suggest the use of dicyanodiamidine for the simultaneousdetection of nickel and cobalt. The former metal gives a crystal-line precipitate,22 and the latter an intense reddish-violet colora-tion.23W. Neumann 24 describes an electrolytic method whereby 0.008milligram of zinc in 0.1 C.C. of solution may be detected. L. W.McKay 25 draws attention to the danger of zinc sulphide re-dissolvingwhen precipitated in presence of sodium hydroxide.Compt. rend., 1908, 146, 628 ; A., ii, 430.lo Boll. chirn. farm., 1908, 47, 73 ; A., ii, 322.l1 Zeitsch. anal. Chew., 1908, 47, 690 ; A., ii, 984.l2 Analyst, 19d8, 33, 450 ; A ., 1909, ii, 95.l 3 Compt. rend., 1907, 145, 697 ; A., 1907, ii, 989.l5 B e y . , 1905, 38, 2520 ; A., 1905, ii, 613 ; Kraut, Zeitsch. angew. Chem., 1906,l6 Ann. Chim. anal., 1908, 13, 16 ; A., ii, 133.l7 Ibid,, 390 ; A . , ii, 988.l9 Bcr., 1906, 39, 3356 ; A., 1906, ii, 908.2o See, fnrther, Chem. Zeit., 1908, 32, 804 ; BUZZ. SOC. chin?., 1908, [ivl, 3, 894 ;21 Ber., 1908, 41, 1878 ; A., ii, 635. Ann. Report, 1907, 205.23 See also H. Grossmann, Chin. Zeit., 2908, 32, 315 ; A . , ii, 434.24 Zeitsch. Elcktrochcm., 1907, 13, 751 : A., ii, 67.Ann. &port, 1907, 200.19, 793 ; A., 1906, ii, 858 ; Brunck, Ann. Report, 1907, 205.Bull. SOC. chim., 1908, [iv], 3, 14 ; A., ii, 230.A . , ii, 899.J. Amer. Chem. Soc., 1908, 30, 376 ; A., ii, 431ANALYTICAL CHEMISTRY.183W. Bette126 states that molybdic acid gives, with hydrogenperoxide and a trace of ammonia, a brownish-red coloration. Amethod for the detection of ruthenium in platinum alloys bas beendevised by N. A. O r l ~ f f . ~ ~N. Schoor128 describes the appearance under the microscope ofsilver, lead, and mercurous chlorides, and later 20 that of arsenic,antimony, and tin compounds. Subsequently30 he deals with themicrochemical analysis of the sulphides of mercury, bismuth, lead,copper, and cadmium. In connexion with the well-known volatilityof mercury compounds, K. Kof and H. Haehn31 state that amoistened filter-paper placed over a 2 per cent. solution of mer-curic chloride for sixty-five hours is rendered black on treatmentwith hydrogen sulphide; also that a distinct white patch is obtainedon developing a photographic plate which has been kept for twenty-four hours a t a distance of 2 to 3 mm.from a drop of it 0.01 per cent.solution of mercuric chloride.32 J. Moir33 gives two methods forthe detection of mercuric chloride in nitrocellulose.According to M. DelBpine,34 one part of copper in 1,000,000 canbe detected by the brown coloration produced with a solution ofa dialkyldithiocarbamate. Iron gives a pink colour with thisreagent; nickel and cobalt also give colours. E. Enechts5 showsthat when titanous sulphate is added to a solution of a copper salt(limit one part of copper per 1 ,OOO,OOO), metallic copper separates.A. W. Gregory36 states that 0.01 milligram of silver may bedetected by the brown colour produced on addition of ammoniumsalicylate and persulphate.W. J. Earslake’s method for the detec-tion of manganese and chromium in mixtures37 consists in boilingthe solution in nitric or sulphuric acid with ammonium persulphateand silver nitrate, when permanganate and perchromate are formed.When shaken with hydrogen peroxide and ether, the former isdecomposed, whilst the latter dissolves in the ether with the pro-duction of the well-known blue colour.H. Caron and D. Raqueta point out that in the well-knownmethod of detecting barium in presence of strontium and calcium,*6 Chem. News, 1908, 97, 40; A . , ii, 230.27 Chernz. Zeit., 1908, 32, 77 ; A . , ii, 231.?8 Zeikch. anal. Cltem., 1908, 47, 209 ; A , , ii, 432.29 Bid., 367 ; A., ii, 777.30 Bid., 729 ; A , , 1909, ii, 96.31 Arch.Pharnt., 1907, 245, 529 ; A., ii, 69.32 Compare Zeitsch. physikal. Chem., 1907, 60, 367 ; A., 1907, ii, 732.33 Chem. News, 1908, 97, 133.34 Bull. SOC. chim., 1908, [iv], 3, 652 ; A., ii, 633.37 J. Amor. Cl~cnz. Soe., 1908, 30, 905 ; A . , ii, 635.38 Bidl. SOC. chim., 1908, [iv], 3, 453 ; A . , ii, 496.Rer., 1908, 41, 498 ; A . , ii, 270. 36 Proc., 1908, 24, 125184 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.the addition of hydrofluosilicic acid sometimes leads to the deposi-tion of gelatinous silica, which may be mistaken for bariumsilicofluoride. They show 39 that barium may be eliminated fromstrontium salts by fractional precipitation by alkali chromates.J. Milbauer40 states that neodymium gives a colourless boraxbead in the oxidising, and a violet in the reducing, flame, whilstpraseodymium gives a yellowish-green bead in the oxidising, anda green in the reducing, flame. 0.Lutz41 points out that boraxbeads are to be preferred for the detection of positive, and phosphatebaads for negative, ions. J. Donau 42 describes a delicate methodof detecting gold, platinum, and silver by means of the meta-phosphate bead.For the detection of phosphates in minerals, A. P. Lidoff 43heats the finely-powdered mineral first alone, and then with mag-nesium or a mixture of it and aluminium; the fusion is gentlywarmed with 20 per cent. sodium hydroxide solution, the evolvedgases being brought in contact with copper sulphate paper.Theformation of copper phosphide (black coloration) indicates thepresence of phosphorus. J. S. Jamieson4' describes a test forbromides.Quantitative.-An important contribution to the theory ofindicators is that of J. T. Hewitt,45 who points out the insufficiencyof the ionic theory alone in explaining the colour changes ofphenolphthalein, and concludes that the assumption of tautomericchange must also be made. In the light of A. G. Green's experi-m e n t ~ , ~ ~ Hewitt assumes that this and like indicators obey thefollowing equilibrium in solution :O:X,*H Z X,*O*H Z X,*O'+H',where Xu and X, are isomeric radicles. These considerations Zedhim to recommend p-nitrobenzeneazo-a-naphthol as an indicator.It yields results similar to phenolphthalein when used for the titra-tion of alkali hydroxides with weak acids, the colour change beingfrom purple 150 yellow.The view that phenolphthalein exhibitstautomerism was put forward by S. F. Acree in 1904, but wasdenied by J. Stieglitz. Acree's latest hypothesis 47 commends itself,however, t o Stieglitz.48 A. Hantzsch and F. Hilscher49 bring39 Bull. Soc. chim., 1908, [iv], 3, 493 ; A . , ii, 496.4u Zcitsch. anal. Chem., 1907, 46, 657 ; A . , ii, 70.41 It&?., 1908, 47, 1 ; A., ii, 226.42 ZeiLwh. Chem. Ind. Kolloide, 1908, 2, 273 ; A . , ii, 434.43 J. .Buss. Pl~y.?. Chem. Soc., 1908, 40, 817 ; A., ii, 894.44 Proc., 1908, 24, 144.46 Ber., 1907, 40, 3724 ; J. Soc. Chem. Ind., 1908, 27, 4 ; A . , 1907, i, 933.J5 Amer.Chew. J., 1908, 39, 528, 649 ; A . , i, 422, 653.49 Ber., 1908, 41, 1187; A., i, 469.45 A?LaZ~jst, 1908, 33, 85 ; A . , i, 269.Bid., 651 ; A . , i, 652 ; compare Ibid., 789 ; A . , i, 653ANALYTICAL CHEMISTRY. 185forward evidence that helianthin, in the solid state, has a quinonoidstructure, but that in aqueous solution it exists as an aminoazo- andsulphonic acid-form in equilibrium, whilst its sodium salt (methyl-orange) is a sulphonate both in the solid state and in solution.J. H, Hildebrand50 has applied the Konig spectrophotometer tothe measurement of the dissociation c o d a n t of phenolphthaleindissolved in aqueous solutions of ammonia and ammonium chloride,in which the concentration of the hydrogen ions is known.It isfound to be 1*7x10-10 for solutions in which 5 to 65 per cent. ofthe phenolphthalein is dissociated.51 M. Barberio 52 has describe?a new indicator, “ resorubin,” obtained by the action of nitrous acidon resorciiiol ; the violet neutral solution becomes blue with alkalisor yellow with acids. It is said to resemble lacrnoid, but to bemore sensitive in presence of ammonium salts. E. Rupp and R.Loose 53 propose pdimethylaminoazobenzene-o-carboxylic acid as anindicator which may be used for the titration of weak bases suchas alkaloids; it can be used to titrate ammonia even in N/100-solution. To distinguish mineral from organic acids, E. Linder 54employs metanil-yellow paper, which (with the former only) becomesviolet. A. B. Lyons 65 recommends haematoxylin for titrating phos-phoric acid.J. K. Wood 56 has carefully determined the basic andacidic constants of arsenious and aluminium hydroxide. V. H.Veley’s paper on the affinity constants of bases, as determined bythe aid of rnethyl-~range,~’ is of interest to analytical chemists.S. P. L. Sorensen and A. C. Andersen58 give some useful hintson Winkler’s method of estimating hydroxides in presence of car-bonates.69 L. Clarke and C. L. Jackson60 show that rosocyanin,the substance which is produced in the test for boric acid withturmeric paper, is an isomeride of curcumin, Cl4HI4O2.I n connexion with halogen derivatives, P. Jannasch 61 has con-tinued his studies on the separation of the halogens by means ofhydrogen peroxide.62 I n its present form, his method is quantita-tive, good results being obtained for chlorine and iodine, whilst thosefor bromine are slightly low.H. Baubigny 63 has modified Hager’s50 Zeitsch 3lektrochem., 1908, 14, 349 ; A., ii, 646.51 Compare R. Wegscheider and A. Schugowitsch, Jblbid., 510 ; A , , ii, 806.52 Gaxzetta, 1907, 37, ii, 577 ; A., i, 161.53 Ber., 1908, 41, 3905 ; A., 1909, ii, 90.54 J, Soc. Chem. Ind., 1908, 27, 485 ; A , , ii, 627.55 Phnrm. Bev., 1908, 26, 9 7 ; A., ii, 532.56: Trans., 1908, 93, 411.57 Ibid., 652.59 Compare also Anclersen, Tihkr. Kem. Farm. Terapi, 1908, 11, 161 ; A., ii,( j l J. pi’. Chcm., 1908, [ii], 78, 28 ; A . , ii, 730.54 Zeitsch. anaZ. C?mn., 1908, 47, 279 ; A., ii, 534.Amer. Chem. J., 1908, 39, 696 ; A ., i, 670. 985.Ann. RepoTt, 1906, 203. 63 Compt. rend. 1908, 146, 335 ; A., ii, 321186 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.method of separating silver halides.s4 When only the chloride andiodide are present, these may be separated quantitatively by treat;ment a t 70--80° with a solution containing 10 grams of ammoniumsesquicarbonate and 20 C.C. of 20 per cent. ammonia per litre; thisdissolves the silver iodide only. B. H. Buttle and J. T. Hewitto5have studied the solubility of silver chloride in mercuric nitratesolution, and arrive a t the same conclusion as Morse>@ that whenmercuric nitrate is present in large excess, chlorine only occurs asHgCl ions. For the determination of fluorine in rocks, etc., inquantities up to 3 per cent., G.Steiger67 makes use of the factthat fluorides bleach the coloration produced by hydrogen peroxideand titanium. As this is not a linear function, reference is madeto a curve. The presence of aluminium salts and phosphatesinfluence the results, which, however, are in no case trustworthywhen as much as 10 per cent. of fluorine is present. B. Carlsonand J. Gelhaar 68 deal with the detection and estimation ofchlorites and hypochlorites in chlorates. E. Knecht 69 has deviseda, volumetric method for the estimation of chlorates. D. Venditori 70makes the interesting observation that, in presence of sulphuricacid, chlorates but not perchlorates are reduced by finely-dividedaluminium. Griitzner’s method of determining chlorates andbromates by heating with formaldehyde, nitric acid, and silvernitrate71 has been extended to iodates and periodates by H.Brunner and R.Mellet.72 To prevent loss of free halogen, and toobviate working in a closed vessel, potassium persulphate is addedto the reaction mixture. H. Baubigny 73 shows that an ammoniacalsolution of silver chloride or bromide, when heated with iodicacid, only deposits silver iodide when a temperature of 200° isreached.Use has been made of sodium peroxide by S. W. Parr74 in theestimation of sulphur in pyrites, coal, and indiarubber. Pyritesis mixed with sodium peroxide, potassium chlorate, and benzoicacid, and the mixture ignited in a special bomb; the melt containsthe sulphur as sulphate. A special mixture appears to be required64 Zeitsch.anal. Chcm., 1871, 10, 341.65 Trans., 1908, 93, 1405.66 Zeitsch. physikal. Chem., 1902, 41, 709 ; A., 1903, ii, 12.67 J. Amer. Chem. Soc., 1908, 30, 219 ; A., ii, 426.68 Chem. Zcit., 1908, 32, 604, 633 ; A., ii, 731.69 J. SOC. Ghem. Ind., 1908, 27, 434 ; A., ii, 627.7O Gazzetta, 1907, 37, ii, 383 ; A., ii, 63.7 l Arch. Phamn., 1896, 234, 634 ; A . , 1897,ii) 166.p2 J. pr. Chem., 1908, [ii], 77, 33 ; A., ii, 222.TJ Compt. rend., 1908, 146, 1097 ; A., ii, 577.74 J. Aiiier. Chem. Soc., 1908, 30, 764 ; A., ii, 628ANALYTICAL CHEMISTRY. 187for each substance. Carbon in carborundum may be estimated(as carbonate) by heating the sample in this way with sodiumperoxide and “ boro-magnesium ” mixture.75 The results quotedare satisfactory (see also under Organic Chemistry).E. Jaboulay 76describes a volumetric method for estimating sulphur in steel. Fora detailed study of the various methods for this purpose, M. Orthey’spaper77 may be consulted. According to H. Kiliani,78 when alkalithiosulphate is titrated in alkaline solution with permanganate, as inReinige’s method of estimating iodides,79 sulphate and not tetra-thionate is formed, hence 8 mols. of permanganate require 3 mols.of thiosulphate for decomposition. V. LenhersO shows that in themethod of estimating tellurium by precipitation with hydrazineF1it is an advantage if the solution contains sulphurous acid; theresults are accurate.As usual, a great many papers have appeared during the yeardealing with the estimation of phosphorus and phosphates.F. W.Hinrichsen 82 has confirmed the accuracy of H. Lidholm’s process 83for the estimation of phosphorus in calcium carbide. An accuratemethod for the estimation of phosphorus in phosphor-tin has beendevised by W. Gemmell and S. L. Archbutt.84. F. Repiton85describes a modification of Malot’s method of estimating phosphoricacid by titration with uranic solution, employing cochineal, in sit@,as indicator; the results are shown to be satisfactory. For theestimation of phosphorus in iron, etc., G. Chesneau 86 weighs theammonium phosphomolybdate precipitate, and states conditionswhereby this compound may be precipitated in presence ofammonium nitrate containing constantly 1.6 per cent. of phos-phorus.With reference to the estimation of phosphoric acid asammonium phosphomlybdate,87 G. von Knorre,88 adverting to hisprevious statement that tungstic acid may be separated from phos-phoric acid almost quantitatively by precipitation with benzidinehydrochloride, points out that the results are vitiated by the insolu-75 LOG. cit.77 Zeitsch. angcw. Chem., 1908, 21, 1359, 1393 ; A . , ii, 731.78 Chem. Zeit., 1908, 32, 1018 ; A., ii, 982i9 Zeitsch. anal. Chem., 1870, 9, 39.go J. Amer. Chem. Soc., 1908, 30, 388 ; A., ii, 426.a1 Gutbier, Ber., 1901, 34, 2724 ; A., 1901, ii, 687.a2 M<tt. K. illaterialpriifs-Amt. Gross. Lichterfelde West, 1907, 25, 110 ; A .BS Zeitsch. angew. Chesn., 1904, 17, 1452 ; A., 1904, ii, 776.84 J. Xoc. Chem. lnd., 1908, 27, 427 ; -4.) ii, 629.85 Mon.Sci., 1907, [iv], 21, ii, 753, 815 ; A . , ii, 320, 428.bG Compt. rend., 1908, 146, 758 ; A . , ii, 427.s7 Compare P. Christensen, Zeitsch. nnnl. Chent., 1908, 47, 529 ; A., ii, 895 ;76 Rev. gen. chim. pure appl., 1907, 10, 193 ; A., ii, 223.1908, ii, 131.E. Rabcn, ibid., 546 ; A., ii, 896. Ibid., 37 ; A., ii, 231188 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.bility of benzidine phosphate in water, but he shows that theseparation is quantitative when the homologous tolidine hydro-chloride is employed.In connexion with the estimation of carbon in iron and steel,M. Orthey89 shows that the combustion method in a current ofoxygen in presence of bismuth oxide gives good results. L. L. deKoninck and E.von Winiwarterm propose to burn the substancewith lead borate in a current of oxygen. H. Isham and J. Aumer 31show that when iron or steel is ignited in a current of oxygen alone,the carbon (but not the sulphur) is almost completely burnt.C. M. Johnson 92 describes an electric furnace for the estimation ofcarbon in iron, etc. New apparatus for the moist method have beendevised by M. Widemann93 and by T. Grzeschik.94 E. P. Mooreand 5. W. Baing5 show that during the solution of steel in potass-ium cupric chloride there may be a loss of 0-4-0.5 milligram ofcarbon per gram of steel. I n titrating lead with sodium sulphide,H. Koch 96 adds carbon tetrachloride to cause the subsidence of theprecipitate. Some useful data on the colorimetric estimation of leadas sulphide are given by H. W.Woudstra.97 Several papers dealingwith the assay of red lead have been published during the year.gsThe last-mentioned shows the limit of accuracy of the various methodsin use. Some useful volumetric methods of estimating mercury aredescribed by E. Rupp,99 and by the same author in conjunction withW. 3'. Schirmer.1 For the estimation of mercuric salts, S. G.Liversedge suggests conversion into mercuric iodide and extractionof the latter with ether; the method is particularly adapted for theestimation of small quantities of mercury. In connexion with theestimation of mercury, the volatility of its salts (see p. 183) is, as arule, ignored by authors. J. F. Spencer and Miss M. Le Pla havedevised a very accurate method of estimating silver and thalliumyQ Chefit.Zcit., 1908, 32, 31 ; A , ii, 131.go Rd1. SOC. chim. Bclg., 1908, 22, 104 ; A . , ii, 320.g1 J. Anier. CJzenz. Soc., 1908, 30, 1236 ; A . , ii, 898.93 Ibid., 773 ; A . , ii, 630.g3 Zeitseh. chem. Apparatenkunde, 1908, 3, 296 ; A., ii, 954.g4 Chem. Zcit., 1908, 32, 1092; A., ii, 1071.g5 J. SOC. Chem. Id., 1908, 27, 845 ; A., ii, 899.9s Chem. Zeit., 1908, 32, 124 ; A . , ii, 227.97 Zeitsch. anorg. Chem., 1908, 58, 168 ; A . , ii, 633.98 See for instance J. F. Sacher, Chem. Zeit., 1908, 32, 62 ; A., ii, 228 ;E. Pieszczek, Pharm. Zeit., 1908, 53, 87 ; A., ii, 228 ; E. E. Dunlap, J. Amer.Chem. Xoc., 1908, 30, 611 ; A . , ii, 537; P. Beck, Zeitsch. anal. Chcm., 1908, 47,465 ; A . , ii, 777.Chem.Zeit., 1908, 32, 1077 ; A., ii, 1073.Pharm. Zeit., 1908, 53, 928 ; A . , ii, 1073.Analyst, 1908, 33, 217 ; A . , ii, 634. ;i 2'raiu., 1908, 93, 858ANALYTICAL CHEMISTRY. 189in mixtures. W. R. Lang and J. 0. Woodhouse4 describe a’modification of Lang and Allen’s apparatus which may be employedfor the estimation of silver by Gay-Lussac’s method.G. S. Jamieson, H. L. Levy, and H. L. Wells6 propose a volu-metric method for the estimation of copper, which, on their evidence,has an average limit of accuracy of one in 300. The copper isprecipitated as cuprous thiocyanate, and the latter, after washing,dissolved in dilute hydrochloric acid and titrated with potassiumiodate and chloroform. The writer can fully confirm H. Theodor’sstatements 7 regarding the accuracy of Volhard’s titration methodof estimating copper; more than twenty years ago he used themethod of neutralising the solution with a slight excess of ammoniabefore reducing with sulphurous acid for the purpose of gettingrid of the nitric acid, which is now suggested by 0.Kuhn.8 A. K.Huntington and C. H. Deschg deal with the planimetric analysisof alloys and the structure of phosphor-copper.The fact that when titrated with permanganate in presence ofhydrochloric acid ferrous salts require more of the standard solutionthan is necessary for their oxidation to the ferric state has longbeen known. T. W. Harrison and F. M. Perkin lo find that additionof manganous sulphate retards the reducing action of hydrochloricacid, but that the colour interferes with the end point of thetitration. In discussing their paper, 0.Heher11 quoted Fre-senius’s directions, and cited a paper by Loewenthal and Lenssen.12In a lengthy paper by L. Brandt,l3 the necessity is urged ofstandardising the permanganate in presence of hydrochloric acidwhen that acid occurs in the assay liquid. Directions are givenfor the preparation of ferric oxide in a state of purity, which isrecommended as the best compound to use in standardising thesolution; the reduction is effected by stannous chloride. In otherpapers on the permanganate method, M. 31. P. Muir14 proposest o arrest the evolution of hydrogen, when the reduction of ferricsalt is accomplished with zinc, by the addition of mercuric chloride;and H.D. Newton15 employs titanous sulphate as reducing agent,destroying the excess with bismuth oxide. The results quoted areTrans., 1908, 93, 1037.J. Amer. Chem. Soc., 1908, 30, 760 ; A . , ii, 634.Chem. Zeit., 1908, 32, 889 ; A . , ii, 898.Trans. Faraday SOC., 1908, 4, 51 ; A., ii, 846.Ibid., 1907, 91, 1370.* Ibid., 1056 ; A . , ii, 1072.lo Analyst, 1908, 33, 43 ; A., ii, 228. 11 Ilk?.l2 Zeitsch. anal. Chem., 1861, 1, 329, 361.l 3 Chem. Zeit., 1908, 32, 812 ; A., ii, 899.Chem. News, 1908, 97, 57 ; A., ii, 228.lr, Amcr. J. Sci., 1908, [iv], 25, 343 ; A., ii, 538190 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.very accurate. G. Edgar16 deals with the estimation of iron andvanadium in mixtures by titration with permanganate, and inanother paper 17 he has elaborated conditions whereby vanadic andmolybdic oxides may be estimated by permanganate.S. B. Jatar'smethod 18 for the estimation of iron and chromium by titration withtitanous chloride appears useful and accurate. H. Bollenbach l9proposed to titrate ferric salts with sodium hyposulphite (Na2S02).For the decomposition of ferro-compounds, especially ferro-silicon,P. Nicolardot2O heats with sulphur chloride. R. B. Moore andI. Miller21 show that iron may be precipitated from it solutioncontaining free hydrochloric acid by means of pyridine ; aluminium,chromium, and zinc are partly precipitated, whereas manganese,nickel, and cobalt remain in solution. F. C. Mathers22 shows thattraces of iron may be removed from indium by precipitation froman acetic acid solution by nitroso-&naphthol.According to A.W. GregoryF3 one part of iron in 10,000 ofcopper may be detected by the red colour given by ferric salts withsalicylic acid in presence of sodium acetate. He bases a colori-metric method on this reaction.With reference to the dimethylglyoxime and dicyanodiamidinemethods of estimating nicke1,24 0. Brunck 25 brings forward evidenceshowing the superior accuracy of the former, whilst Grossmann andSchiick 20 uphold the dicyanodiamidine method on the ground thatthe nickel compound is insoluble in strong ammoniacal alkalihydroxide solutions. H. Cantoni and M. Rosenstein 27 propose forthe volumetric estimation of nickel, titration with either ferro- orferri-cyanide, using ferric (or uranium) or ferrous icdicators' respec-tively; the ferricyanide method with the liquid faintly acidifiedwith acetic acid gives the better results.In a series of papers M.E. Pozzi-Escot deals with the estimationof both nickel and cobalt by the molybdate method and the separa-tion of these metals from others.2sl6 Zeitsch. anorg. Chem., 1908, 59, 74.l 7 Anzer. J. Sci., 1908, [iv], 25, 232 ; A., ii, 540.J. Xoc. Chem. Ind., 1908, 27, 673 ; A., ii, 778.l9 Chem. Zeit., 1908, 32, 146 ; A . , ii, 229.20 Compt. rcnd., 1908, 147, 676 ; A., ii, 1074.21 J. Amcr. Chem. SOC., 1908, 30, 593 ; A., ii, 434.22 Ibid., 209 ; A., ii, 434.23 Trans., 1908, 93, 93.25 Zeitsch. angew. Chem., 1907, 20, 1845 ; A., 1907, ii, 989.26 lbid., 1981 ; A., ii, 71.28 Ann.Chim. anal., 1908, 13, 66 ; Compt. rend., 1907, 145, 1334 ; A., ii, 229 ;Ann. C h h . n?:ccl., 1908, 13, 89 ; A., ii, 324 ; ibid., 85, 215, 217 ; A,, ii, 539,640, 635.24 Ann. Report, 1907, 205.Bull. Soc. chirn., 1908, [iv], 1, 1163 ; A., ii, 230ANALYTICAL CHEMISTRY. 191E. D. Campbell and W. Arthur29 describe a modification ofMoore’s volumetric method 30 for the estimation of nickel andchromium in steel. A. A. Blair 31 has worked out a scheme for thedetermination of vanadium, molybdenum, chromium, and nickel insteel; from the single example given, it appears to be very accurate.The estimation of vanadium in iron and steel is dealt with by E. D.Campbell and E. L. W0odhams.3~In connexion with the estimation of tin, D.B. DottB drawsattention to the solubility of metastannic acid in hydrochloricacid; he points out further that, in the analysis of ores, tin isvolatilised when a solution in aqua regia is evaporated to drynessfor the purpose of separating the silica.34 H. Reynolds35 titratesa stannous solution with dichromate, using azobenzenesulphonic acidas indicator; the restoration of the red colour of the indicatormarks the end of the oxidation. E. Schurmann and W. Scharfen-berg36 describe a modification of Clarke’s oxalic acid method forthe analysis of white metal. A. eKolb and R. Formhals37 showthat the reaction Sb,O, + 4HI f Sb,O, + 2H,O + 21, is practicallycomplete from left t o right if aufficient excess of hydriodic acid beused.L.Rosenthaler’s observations 38 respecting the quantitative pre-cipitation of arsenious and arsenic acids, the former by bariumchloride and ammonia and the latter by barium chloride andsodium hydroxide, are worthy of attention. H. Reckleben and G.Lockemaun 39 describe gravimetric, volumetric, and gasometricmethods of estimating arsine in air.P. Cazeneuve40 utters a warning against the use of arseniferousdressings as insecticides in agriculture. The Gutzeit test for arsenic,antimony, and phosphorus has been rendered more conclusive byB. Sjollema.41 I n the case of arsenic, he states that he was ableto obtain microscopical crystals with 0.005 milligram of arseniousoxide. C. H. Nieuwland,& on the other hand, was unable to2~ J. Amer.Chem. SOC., 1908, 30, 1116 ; A , , ii, 779.3O Chcm. News, 1895, 72, 92 ; A., 1895, ii, 534.31 J. Amcr. Chew. SOC., 1908, 30, 1229 ; A., ii, 900.3 3 Ibid., 1233 ; A , , ii, 901.33 PJzawn. J., 1908, [iv], 27, 486 ; A . , ii, 899.See further, Ibid., 585 ; A . , ii, 1075.35 Chem. News, 1908, 97, 13 ; A., ii, 134.36 Mitt. K. ~nterialprufgs.-Amt., 1908, 25, 270 ; A . , ii, 537.37 Zeitsch. anorg. Chem., 1908, 58, 189 ; A., ii, 599.a* Apoth. Zeit., 1907, 22, 982 ; A., ii, 322.39 Zeitsch. anal. Chem., 1908, 47, 126 ; A . , ii, 224.*O Rev. intern. Palsq., 1908, 21, 11.41 Chenz. WeekblacE., 1908, 5, 11 ; A . , ii, 431.42 Ibid., 558 ; A . , ii, 896192 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.detect less than 0.05 milligram of arsenious oxide in this way.43W.van Ryn44 points out that sodium fluoride may prevent com-pletely the formation of arsenical mirrors in the Marsh-Berzeliusmethod, although it has no influence on the Gutzeit or Fluckigerreaction. E. Salkowski 45 deals with the detection and estimation ofarsenic in urine. To detect arsenic in sulphur, J. Brand 46 extractsthe finely-powdered sulphur with ammonia.M. Orthey47 has made it careful study of certain volumetricmethods of estimating manganese in ferro-manganese ores, and heshows that the Volhard-WolE method 48 and those of von Knorre 49and Blair give results agreeing well with those yielded by theordinary gravimetric method ; the first two are specially recom-mended.51 For the detection and colorimetric estimation of man-ganese, M.Duyk52 makes use of the red coloration produced whena trace of manganese is heated with an alkaline solution ofhypochlorite in presence of a trace of copper sulphate. H. W.Rowel1 53 estimates small quantities of bismuth colorimetrically asiodide.Miss Z. Kahan54 has devised a method for the quantitativeseparation of barium and strontium. 0. Hauser, in conjunctionwith F. Wirth, has published some important data on the solu-bility of the oxalates of the rare earths in dilute sulphuric andoxalic acids.55 Later 56 he shows that the precipitation of theseoxalates is incomplete in presence of uranyl salts. M. Dittrich67describes the separation of cerium from other metals and itsestimation.To separate lithium from other alkali metals, L.Eahlenberg andF. C. Krauskopfbs take advantage of the solubility of lithiumchloride in pyridine. The precipitation of potassium as cobalti-45 For microchemical reactions of arsenic, see also G. Denigks, Contpt. Tend.,1908, 147, 596, 744 ; 8., ii, 1070.l4 Phnrm. Weekblnd., 1908, 46, 98 ; A . , ii, 224.4.i Zeitsch. physiol. Chem., 1908, 56, 95 ; A., ii, 734.46 Zeilsch. ges. Brauwesen, 1908, 31, 33 ; A., ii, 532.47 Zeitsch. anal. CJzem., 1908, 47, 547 ; A., ii, 898.48 Ihid., 1880, 20, 271 ; compare Volhard, A , , 1880, 141.49 Ibbid., 1904, 43, 643 ; compare A., 1902, ii, 108.Ibid., 1904, 43, 647 ; compare A., 1904, ii, 683.51 Compare E. W. Meyer, Zeitsch. angew. Cheni., 1907, 20, 1980; A., ii, 71 ;52 Ann. Chim. anal., 1907, 12, 465 ; A., ii, 70.63 J.Soc. Chem. Ind., 1908, 27, 102 ; A., ii, 325.64 Analyst, 1908, 33, 12 ; A, ii, 133.55 Zeilsch. anal. Chem., 1908, 47, 389 ; A., ii, 778.56 Ibid., 677 ; A., ii, 987..57 Ber., 1908, 41, 4373.58 J. Amcr. Chem. Xoc., 19OS, 30, 1104 ; A., ii, 777.L. Sacerdoti, L’lndustria Chimica, 1907, 7, 258 ; A., ii, 228ANALY'I'ICAL CHEMIS'I'HY. 193nitrites9 is dealt with by W. A. Drushe1,GO and by W. Auteiirieth,61who shows that the precipitate obtained with de Koninck's cobaltreagent is not of constant composition.For the estimation of tungsten and its separation from othersubstances, advantage may be taken of the fact that it is volatilisedas oxychlorides when heated strongly in a mixture of chlorine andsulphur chloride,62 or when heated at 500° in a current of aircharged with chlorine.63 Tungsten trioxide is reduced by hydrogenat 600-900°, and may be then volatilised by treatment withchlorine.64 G.von Knorre has applied his benzidine method@ tothe estimation of tungsten in steel containing chromium.GG F. W.Hinrichsen and L. WolterG7 state that the results are low. Theydescribe other methods of estimating both tungsten and chromium.Methods for the estimation of vanadium in presence or absence ofiron are described by T. Warynski and B. Mdivani,G8 and byG. Edgar.69Electrochemical A naZysis.-The work conducted in this depart-ment during the year indicates steady progress, and shows theincreasing utility of electrolytic methods. F. Foerster 70 summarisesour knowledge of rapid electrolytic methods, especially those inwhich rotating electrodes are employed.His claim of priority forrotating electrodes has been disputed by A. Classen,71 and a seriesof polemical papers by these authors have followed.72 F. M.Perkin 73 remarks that rotating electrodes were first described byGooch and Medway, E. Smith, and himself almost simultaneously.F. A. Gooch and F. B. Beyer 74 employ as cathode a Gooch cruciblewith asbestos filter in the case of precipitates which do not adherefirmly.The reduction of alkali nitrate to ammonia in presence of copperwas studied by Easton in 1904, and subsequently by Ingham in1905. 0. L. Shinn75 states that in order to realise Ingham's5g Adie and Wood, Trans., 1900, 77, 1076.G1 Centr.Min., 1908, 513 ; d., ii, 897.tj2 F. Bourion, C'onzpt. T C ~ Z ~ . , 1908, 146, 1102 ; - 4 . , ii, 737.63 P. Nicolardot, i b i d , , 147, 795; A . , ii, 1074.64 E. Defacqz, ibid., 146, 1319 ; d., ii, 737.G Ann. h'epwt, 1905, 192.Zdsch. nital. Chenz., 1908, 47, 337 ; A . , ii, 779.67 Zeitsch. nnorg. Chem., 1908, 59, 183 ; A., ii, 900.Ann. Chint. anal., 1908, 13, 209, 210 ; A., ii, 636, 736.69 Amer. J. Sci., 1908, [iv], 26, 79 ; A . , ii, 736.7O Zeitsch. Eleklroehent., 1908, 14, 3.i2 Ibid., 141; 208, 239; A . , ii, 432, 529.76 Amer. J. ,Ski., 1908, 25, 249 ; d., ii, 529.75 J. Amm. Chem. SOC., 1908, 30, 1378 ; A . , ii, 893.Zeitsch. cmorg. Chein., 1907, 56, 223 ; 1898, 59, 97 : A., ii, 66, 735.71 Ibid., 90.73 Ibid., 143; A , , ii, 432.REP.-VOL.V. 194 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.results, the anode must be rotated slowly; the current should be4-5 amperes a t 10 volts, and not more than 20-25 C.C. ofN/5-sulphuric acid should be present. He also finds that it issometimes necessary to add a second or even a third quantity ofcopper sulphate in order to complete the reduction.H. W. Gillett 7G describes the conditions necessary for the deposi-tion of silver in presence of copper from ammoniacal tartratesolutions ; an important point is that when the electrolysis is carriedout at a temperature below 20° very rapid rotation of the anode isnecessary. Miss M. E. Holmes77 deals with the separation ofcadmium from a large number of metals. 0. Scheen78 gives somedetails for the electrolytic estimation of antimony.I n conjunctionwith his paper should be read that of E. C0hen.7~ H. 5. S. Sandhas continued his useful studies on the rapid separation of metals,POand in a second communication81 it is shown that by means of theapparatus previously described with rotating anodes, coherentdeposits of antimony may be obtained from solutions in sulphuricacid (1 : 1) in the presence of hydrazine sulphate; the temperaturemust be kept above looo with a cathode potential measured againstan auxiliary 2N-sulphuric acid of 0.65 volt. Tin is not depositedunless the auxiliary potential rises above 0.8 volt. I n order toeffect its deposition after separating the antimony, oxalic acid isadded, the solution neutralised with ammonia, and again acidifiedwith sulphuric acid.Considering the difficulty of this separationon theoretical grounds, the results are extremely good. For theseparation of copper from iron, an apparatus is described, includinga parchment paper diaphragm. The influence of temperature onthe estimation of copper is discussed by 5. R. \Vithrow,B2 whilstE. E. Free 53 deals with the estimation of small quantities ofcopper. A. Thiels4 shows that nickel can be estimated accuratelyby the electrolysis of the nitrate provided no nitrite be present,sufficient excess of animonia be added, and a straight wire ofpassive iron used as anode (platinum anodes are attacked); hegives conditions for the separation of nickel and copper quantita-tively. F. Foerster and W.Treadwell 55 confirm Thiel and Windel-76 J. Physical Chem., 1908, 12, 26 ; A . , ii, 226.77 J. Amer. Chem. Xoc., 1908, 30, 1865.78 Zeitseh. Elektrochein., 1908, 14, 257 ; A., 5, 636.Ilbid., 301 ; A . , ii, 636.f o A m . ZZeport, 1907, 207.s1 T r a 7 ~ ~ . , 1908, 93, 1672.g2 J. Amer. CJwa. Soc., 1908, 30, 381 ; A., ii, 432.a; J. Physical Chcm., 1908, 12, 25 ; A . , ii, 227. -s4 Zciitsch. Elektt*ochena., 1908, Id, 201 ; A., ii, 535.S9 Ibid., 89 ; A., ii, 321ANALYTICAL CHEMISTRY. 195schmidt’s statement,*$‘ that in separating nickel from zinc by elec-trolysing an ammoniacal solution containing sodium sulphite aselectrolyte, the deposited nickel contains sulphur. As pointed outby A. Fischer, h0wever,~7 without sodium sulphite a much highercurrent density is required.He gives conditions for the estimationof the two metals. It would appear advisable, taking into accountthe observations of the various authors, to dissolve the nickel firstdeposited and again electrolyse. In the electrolytic estimation ofnickel, A. Schumann shows 88 that a gauze cathode is preferable to adisk or cone cathode. R. Goldschmidt,89 in the electrolysis of zincsilicofluoride, uses a stationary slanting anode, over which thesolution passes and is returned by a pump. F. J. Metzger andH. T. Beans’ method of estimating bismuth by electrolysing anacetic acid solution seems to be very accurate on the evidencegiven.go J. PesettY91 after electrolysing a bismuth salt, adds a knownweight of cadmium sulphate, and continues the electrolysis.Thedeposited cadmium is said to protect the bismuth from oxidationand to render it more adherent.G. Gallo and G. Cenni92 state that by electrolysing a solutionof thallium sulphate, faintly acidified with oxalic acid, at theordinary temperature in a Classen’s capsule with a rotatingplatinum cathode, the whole of the thallium is deposited a t theanode apparently as a new oxide, T1,0,.J. S. Goldbaum and E. P. Smith 93 have continued their experi-ments on the separation of the alkali metals95 to the chloridesof ammonium, caesium, rubidium, and lithium with satisfactoryresults ; attempts to separate potassium and ammonium wereunsuccessful.General.-Sir W. Crookes95 points out the utility of iridiumcrucibles in analysis, chiefly on account of the high resistance ofthe metal against the attack of reagents; but he states that rhodiumpossesses almost the same resistance, and its cost would be lessbecause of its lower density.It is interesting to note that fusedsilica apparatus, which is adapted for so many purposes in analysis,has recently been considerably cheapened.8fi878s899Q9192939195Zeitsch. aizgew. Cl~em., 1907, 20, 1137 ; compare A., 1907, ii, 601.Chem Zeit., 1908, 32, 185 ; A., ii, 324.Zeitsch. ungezu. Chem., 190e, 21, 2570 ; A . , 1909, ii, 97.Bull. SOC. chim. Bely., 1908, 22, 138 ; A., ii, 536.J. Amer. Chem. Soc., 1908, 30, 589 ; A , , ii, 541.Zeitsch. anal. Chem., 1908, 47, 401 ; A . , ii, 780.Atti R. Acad.Lincei, 1908, [v], 17, ii, 276 ; A . , ii, 986.J. Amer. Chein. SOC., 1908, 30, 1705 ; A . , ii, 1072.Ibid., 1907, 29, 447, 1445, 1455; A., 1907, ii, 574, 988.PTOC. Xoy. SOC., 1908, 80, A., 535 ; A . , ii, 702.0 196 ANNUAL REPORTS ON THE PROGRESS OF CHERIISTRY.A modification of the Ostwald pipette for calibrating burettesand other measuring vessels has been devised by *O. von Spindler.96Other useful apparatus in connexion with volunietric analysis isdescribed by G. Muller97 and G. Miiller and 0. Berchem.98A. Gawalowski 99 has devised an apparatus for mixing liquidsduring a reaction. A simple manometer for use when distill-ing under diminished pressure is described by N. L. Gebhard,land H. Siicht,ing 2 has devised an automatic stirring arrangementwhich can be applied to liquids liable to bump when being distilledunder diminished pressure.P. A. Kober 3 describes an apparatuswhereby Folin’s method may be applied to Kjeldahl nitrogendeterminations; and, later,4 he deals with the estimation of carb-amide by Folin’s method. A new form of pyknometer for veryaccurate work has been devised by W. R. Bousfield.5Gas A naZysis.-Useful methods for the detection of ozone,hydrogen peroxide, and nitrogen peroxide in air are describedby E. H. Keiser and L. McMaster.6 P. Mknihre7 gives a methodfor the estimation of mercury vapour in air, whilst J. Ogier andE. Kohn-Abrest 8 deal with the detection and estimation of smallquantities of carbon monoxide in air. C. A. Keane and H.Burrows 9 show that the autolysator for the automatic determina-tion of carbon dioxide in furnace gases gives accurate results.A.Fraenckel 10 deals with the estimation of phosphorus, sulphur,and silica in acetylene. W. A. Bone and R. V. Wheeler l1 describean apparatus for the analysis of mixtures of hydrocarbon gases,and A. E. Hill12 describes a new gas burette. A. Stock13 dcaIswith the uses of the mercury trough in the manipulation of gasesand their treatment with reagents. He gives some useful hints onthe re-purification of mercury. New gas analysis apparatus hasw SchweQ. W o c h . Chenz. Piiurm., 1908, 46, 145 ; A., ii, 625.97 Chem. Zeit., 1908, a, 5 3 2 ; A,, ii, 626.98 Ibid., 711 ; A., ii, 775.9y Zcitsch. m~nb. Ckem , 1908, 47, 697 ; A . , ii, 939.YP’oc., 1908, 24, 51.Zcitsch.C I I L C C ~ . Chem., 1908, 47, 755 ; A,, 1909, ii, 35.[bid., 1279 ; A., ii, 893.Trans., 1908, 93, 679.Anier. Chem. J., 1908, 39, 96 ; A . , ii, 222.Ann. Cicinz. anal., 1908, 13, 169, 218 ; A . , ii, 631, 632.J. Xoc. Chem I ~ t d . , 1908, 27, 608 ; A., ii, 735.J. Soc. Chcm. Iwd., 1908, 27, 10 ; A . , ii, 221.,?3ei*., 1903, 41, 3834 ; A., 1909, ii, 89.:3 J. Amer. ChenL. S‘OC., 1908, 30, 1131 ; A., ii, 626.7 C’onhpt. reud., 1908, 146, 754 ; A., ii, 433.lo J. Gnsbelezcclrt., 1908, 51, 431 ; A . , ii, 983.l2 Tram., 1908, 93, 1857AXALYTICAL CHEMISTKY. 197been devised by R. Ross and J. P. Leather.14 H. Franzenls hasdevised a simple apparatus by meam of which a gas may beabsorbed from a mixture containing a large proportion of non-absorbable gas.L. M. Dennis and E. S. McCarthy16 recommendan ammoniacal solution of nickel cyanide as an absorbent forbenzene in illuminating gas.lVater AnaZysis.-F. Telle 17 proposes a solution containing0.344 gram of gypsum per litre as a sta'ndard in the Clark process.The method suggested by C. J. Blacher and J. Jacoby,? of estimat-ing alkaline earths by titration with potassium stearate and phenol-phthalein, is worthy of attention. H. Nolll9 gives a method for theestimation of carbonic acid and carbonates in chalybeate waters. E.Ernyei 2o describes a method of estimating manganese in waters.A useful paper on the systematic investigation of potable watersis that by G. Romyn,21 who discusses the flora and fauna of potablewaters.22 J.E. Purvis and R. M. Courtauld23 show that organicnitrogen compounds are attacked to a certain extent by the copper-zinc couple, so that the estimation of nitrates and nitrites by thismethod in presence of organic nitrogen compounds gives too highresults. A simple apparatus for observing the rate of absorptionof oxygen by polluted waters has been devised by W. E. Adeney.Z4I<. Saito25 points out that Bacillus coli comrnunis is so widely dis-seminated that great caution must be exercised in condemning awater as polluted because of its presence.G. 0. Adams and A. W. Kimbal126 show that, in the estimationof nitrogen in sewa,ge by the Kjeldahl method, the ammoniaformed may be nesslerised. H. W. Clark and G. 0. Adams 27 statethat the odour and appearance of incubated effluents give a betteridea of their putrescibility than does the measurement of the timerequired to decolorise dyes.I4 J.SOC. C I m i . I?irZ., 1908. 27, 491 ; A., ii, 626.l5 Zeitsch. n n o ~ g . Chcni., 19OS, 57, 395 ; il., ii, 425.lo ,J. Antcr. Chma. SOC., 1908, 30, 2YS ; il., ii, 435.l 7 J. Plmrm. Chilli., 1908, [vi], 27, 380 ; A . , ii, 63%.'8 C ' l ~ ? r i . Zeit., 1908, 32, 744 ; A . , ii, 807.'Jo Chcm. Zcit., 1908, 32, 41 ; A., ii, 133.F2 C'hein. Weekblnd. Reilnyc, 1908, 3015 ; P h n ~ m l17ccX:blrrtl, 1905, 45, 1287.s Yroe. Camb. Phil. SOC., 1903, 14, 441 ; A., ii, 776.24 Sc7:. PTOC. IZoy. Dub/. ,j'or., 1903, 11, 280 : A . , ii, 781.25 Amh. Byqienr, 19OS, 63, 215.O6 J. A?)LcY.CILcm. ,Yor. 1908, 30, 1033.Zcitsch. nngezo. Chc?n., 1908, 21, 840, 1455 ; (l. T,iiiigt>, ibid., 633 ; A , , ii, 4%.Phnrin. lvcekblad, 1908, 45, 402.2; IOfd., 1037198 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Organic Chemistry.Qualitative.-This portion of the literature is as voluminous asusual, but it will only be necessary to mention a few papers.L. E. Hide128 deals with the detection of methyl alcohol in ethylalcohol. I. Lifschutz29 states that the green colour produced bywarming an acetic acid solution of cholesterol with benzoyl peroxide,although less sensitive than Liebermann’s reaction, has advantagesin the position of the absorption bands. Some colour reactionsof cholesterol and oxycholesterol are described by L. Golodetz,30 and,by the same author,31 a colour reaction with formaldehyde andbenzoyl peroxide.I n connexion with Hehner’s test for formalde-hyde in milk,32 H. D. Dakin33 characterises a number of aliphaticketones and aldehydes by the appearances and melting points oftheir p-nitrophenylhydrazones. H. J. H. Fenton and G. Barr34tabulate the colour reactions produced by formic, oxalic, dihydroxy-tartaric, pyruvic, aa-dimethylglutaric, lactic, saccharic, laevulic, andoxalacetic acids when treated with resorcinol, phenol, pyrogallol,or o-cresol in presence of concentrated sulphuric acid. Accordingto W. M. Dehn and S. F. Scott,35 sodium hypobromite givescharacteristic colours with phenols and aromatic amines. T. Silber-mann and N. Ozorovitz36 make use of the formation of resinouscondensation products (insoluble in ordinary solvents) from formal-dehyde and dihydric phenols to detect and identify the latter.F.A. Steensma’s observation^,^^ that an aromatic or heterocyclicaldehyde in presence of a mineral acid gives a colour with phenolsor with heterocyclic compounds containing the group C:CH, aremost interesting as showing that certain colour reactions canactually be predicted, a generalisation of which he cites specificinstances. The formation of an additive (red) compound withp-dimethylaminobenzaldehyde is recommended by B. von Pawlew-ski38 as a characteristic reaction of anthranilic acid. Some newdifferential reactions of the naphthols are described by Volcy-Boucher.39 C. Lefebvre 40 describes some biochemical reactions2* Analyst, 19OS, 33, 417 ; A ., ii, 1056.3O Chcm Xeit., 1908, 32, 160 ; A . , ii, 328.31 See F. von Fillinger,, Zeitsclz. Ifah?-. Genusmz., 1908, 16, 226 ; 8., ii, 902.3* J. Biol. Chem., 1908, 4, 235 ; A . , , ii, 234.Proc. Canzb. Phil. Soc., 1908, 14, 386 ; A . , ii, 438.3B J. Arne?.. C h m . SOC., 1908, 30, 1418 ; A , , i, 780.36 R i b t . SOC. SC~. B?Leuresei, 1908, 17, 41 ; A., 1909, ii, 9s.:(7 Biochenz. Zeitsclt., 1909, 8, 203 ; A,, ii, 442.3* Ber., 190S; 41, 2353 ; A., i, 638.m Am. C‘/iint. a?inE., 19OS, 13, 335 ; A., ii, 990.4o Arch. Phawi~., 1907, 245, 493 ; A!., ii, 57.2D Ber., 1908, 41: 252; A., ii, 233.31 Jbid., 245 ; A . , ii, 330ANALYTlCAL CHEMISTRY. 199(enzymic) for the detection of sugars and glucosides in the Taxacex,B.Tollens and F. Rorive 41 describe colour and spectral reactions ofsugars and their derivatives on treatment with naphtharesorcinol.B. Tollens42 uses the same reagent for the detection of glycuronicacid in ~ r i n e . ~ 3 E. C. Kendall and H. C. Sherman 44 recommendpbromobenzylhydrazide as a reagent for the detection and identifi-cation of various sugars; the hydrazone obtained from galactose isinsoluble in boiling alcohol, whilst those of mannose and lzvuloseare sparingly, and that of dextrose readily, soluble. It is well knownthat even the hexoses give traces of furfuraldehyde when boiledwith acids. C. Fleig 45 shows that furfuraldehyde gives colorationswith indole and with carbazole. L. Garnier 46 deals with somecolour reactions of digitalis glucosides.In connexion with alkaloids, C.Reichard 47 describes somereactions of tropacocaine, which serve to differentiate it fromcocaine. G. Denig&s48 gives some reactions of hordenine, andL. Krauss 49 some reactions of synthetic suprarenine.F. Schulz50 calls attention to the red coloration given when asolution of (crude, not pure) picric acid in benzene is added to amineral oil, a reaction which serves for the detection of the latter inanimal and vegetable oils. Tests for oleic acid are described byA. Manea 51 and I. Lifschutz.52&ziuntitutiva.-There are a few papers dealing with elementaryanalysis demanding notice. M. Dennstedt 53 details the precautionsto be taken in his simplified method of determining carbon indifficultly combustible substance^.^^ He describes 6j a new form ofsoda-lime absorption apparatus, and A.E. Hill56 a new form ofpotash bulbs. W. Lenz57 points out that the percentage of carbon41 Ber., 1908, 41, 1783 ; A . , ii, 638.43 See also K. Tollens, Zeitsch. physiol. Chmz., 1908, 56, 115 ; A . , ii, 740 ; andcompare 6. A. Mandel and C. Neuberg, Biochcm. Zeitsch., 1908, 13, 148 ; A., ii,993.Ibid., 1788 ; A., ii, 639.44 J. Amer. Chem. Soc., 1908, 30, 145 ; A., ii, 902.45 J. Pharm. Chinz., 1908, [iv], 28, 385 ; A . , ii, 1077.46 Ibid., 27, 369 ; A., ii, 544.47 Plinrm. Zentr.-h., 1908, 49, 337 ; A . , ii, 643.48 Bull. Unoc. chim., 1908, [iv], 3, 786 ; A., i, 735.4g Apoth. Zeit., 1908, 23, 701.51 Bulb.Soc. Sci., Bucuresci, 1908, 17, 256.Ii3 Bcr., 1908, 41, 60 ; A . , ii, 321.54 Compare J. ZtLleski, BdI. 24cad. Sci. Crrccou-., 1907, ii, 646 ; .4., ii, 132.65 CIwn. .%it., 1908, 32, 77 ; A., ii, 225.57 Zeitsch. anal. Chcni., 1907, 46, 557 ; A . , ii, 65.50 Chew. Zeit., 1908, 32, 345.Zcitsch. p12ysiol. Chenl., 1908, 56, 446 ; A . , i, 754.Proc., 1908, 24, 182200 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.obtained by the Carrasco-Plancher method58 is apt to be low onaccount of the formation of carbon monoxide. 0. Carrasco andE. Bellonisg recommend a modification of the method, in whichthe organic substance is mixed with powdered platinised biscuitporcelain as catalyst instead of cupric oxide. M. Dennstedt andF. Hassler 60 describe a method and apparatus for the simultaneousestimation of carbon, hydrogen, halogen, and nitrogen, and fromthe one example given-analysis of pyridine platinochloride-theresults would appear to be accurate enough for purposes of check.61For the estimation of the halogens and sulphur, W.Parr showsthat his method of heating with sodium peroxide and “boro-magnesium ” mixture (see p. 186) may be used. To estimate sulphurin indiarubber, he burns the sample with sodium peroxide, potass-ium chlorate, and benzoic acid.62 According to H. D. Richmond,63the higher results obtained when the nitrogen in casein is estimatedby Dumas’s method, as compared with Kjeldahl’s method, is dueto the presence of unburnt carbon monoxide in the former case.He also describes a simple method of estimating nitrogen in com-pounds containing the triazo-group.A series of papers have beenpublished by P. Blackman 64 on a new method of estimating vapourdensity.J. Herzog and V. H. H h c u 65 show that the number of hydroxylgroups present in a phenol may be determined by condensing itwith diphenylcarbamyl chloride, hydrolysing the resulting urethane,and weighing the diphenylamine formed. A. Kirpa166 points outthat methoxyl determinations in such a substance aseither by the Zeisel or by the Herzig-Meyer method are invariablylow, probably due to the wandering of the methyl group from theoxygen to the nitrogen at0rn.~7G. T. Morgan and T. CookG8 describe a useful distillation flaskadapted for use in many methods for the analysis of organic sub-5R Ai~ii.l k p o ~ t , 1906, 21 1.(io Bdr., 1908, 41, a778 ; A . , ii, 984.61 Coml~tre J . %ellenter, Prograiiim cl. Obcrrcnlschulc, Jii,7isbrucX*, 1908 ; Chcn).w2 See p. 186.64 Bcr., 1908, 41, 768, 881, 1558, 2487, 4141 ; c o i n p t ~ i ~ Pro(:., 1908, 24, S :65 Ber., 1908, 41, 638 ; A . , ii, 327. 66 Ibitl., 819 ; A., ii, 436.(;; Compare J. Werzig, Monntch., 1908, 29: 295 ; A., ii, 638.GJ Analyst, 1908, 33, 118 ; A . , ii, 424.J. Pha~in.. Ch,itii., 1008, [vi], 27, 469 ; A., ii, 631.Zmtr., 1908, ii, 635.Annlyst, 1908, 33, 179 ; A . , ii, 530.A., ii: 157, 564 ; A , , 1909, ii, 21ANALYTICAL CHEMISTRY. 201stances in which distillation is necessary. R. Corradi 69 recom-mends the gasometric (hypobromite) method of determining theammonia formed in the Kjelda,hl method.V. von Cordier70describes a modification of Hiifner's hypobromite method for theestimation of nitrogen, and a special apparatus for carrying it out.H. Bollenbach 71 has modified de Haen's volumetric process ofestimating f errocyanides, and the evidence he brings forward showsthat the results are trustworthy. A valuable paper on the analysisof commercial ferrocyanides is that by H. G. C0lman.7~ The paperdeals with the direct estimation of the ferrocyanide by titrationwith copper or zinc sulphate, the estimation of the iron and thecalculation of the ferrocyanide from the results, and the estimationof the hydrogen cyanide. G. Heikel 73 has published a useful paperon the limits of accuracy of the Messinger and Denig'es' methodsof estimating acetone.V. F. Herr 74 describes a new dephlegmatorfor the fractionation of naphtha. I. Bay75 shows that carbondisulphide may be estimated in benzene by precipitating withphenylhydrazine, and weighing the resulting phenylhydrazinephenylthiocarbazate, CS,(Ph-NH=NH,),, dried in a vacuum desicca-tor. This process is, however, not new; it is commonly used, andwas first described by Liebermann and Seyewetz.76 According toF. U ~ Z , ~ ~ when picric acid is heated with sodium hydroxide andhydrogen peroxide, the whole of the nitrogen is obtained as nitrate,and may be estimated by the " nitron " method, whilst 34. Buschand G. Blume 78 show that picric acid may be estimated by weighingit as " nitron " picrate; halides and their oxygenated salts, nitrates,nitrites, and chromates must be absent.F. Raschig79 gives amethod for estimating m-cresol in presence of the 0- andpisomerides ; it is adversely criticised by J. Rerzog.soThe methods of estimating starch in cereals, which depend ondetermining the reducing power after hydrcjlysis with either diastaseor acid, or with both successively, are lengthy and inaccurate,since reducing sugars other than those originating from the starchare invariably present among the products of hydrolysis. AlthoughRoll. cJiL',ii. fn?'jI)., 1907, 46, 861 : -A!., ii, 139.;" Z i l s c h . nnn?. CJ1e7)7., 1908, 47, 682 ; - I . ? ii, 96:;.]bid., 6 S i ; -1.. ii, 996.j 2 Analyst, 1908, 33, 261.8 . ' Chem.Zezt., 1908, 32, 75 : A , , ii, 235.Ibid., 14s ; A., ii, 232.75 Contpt. rend., 1908, 146, 132 ; A., ii, 2.26.iG Ber., 1891, 24, 788 ; A., 1891, 681.7i Zcitsck. m a l . Chwz.. 1908, 47, 140 ; .4., ii, 233.7s Zcitsch. a7~gew. CJwn., 19OS, 21, 354 ; lf., ii. 328.79 PJiccl-m. Zeit., 19OS, 53, 99; A., ii, 233.-.Zhl'd., 141 ; A., ii, 233202 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.no known method gives anything but approximate accuracy, inthe writer’s hands the most satisfactory and rapid method is theDubrunfaut-Effront, which consists in triturating the ground cerealwith concentrated hydrochloric acid and polarising the solublestarch. proposed a modification of thismethod, and later 0. Wengleins2 suggested the use of sulphuricacid (D 1.7) instead of hydrochloric, under which conditions therotatory power of barley starch is found to be [.ID 191.7O.Lintnerhas confirmed Wenglein’s results,83 and adopts his suggestions.E. Ewers84 points out that substances other than starch are dis-solved, and that the results are thereby vitiated. He proposesmodifications, his latest suggestion 85 being treatment with one percent. hydrochloric acid a t looo, and clarification with ammoniummolybdate, before polarising. I n connexion with sugars, Barfoed’scopper acetate solution was put forward as one which was notreduced by maltose and lactose, and F. C. Hinkel and H. C.Sherman 86 have determined the conditions under which in a mixtureof dextrose and the reducing bioses mentioned, only the formerexhibits reducing power.F. Watts and H. A. Tempany87 recom-mend the use of dry basic lead acetate for clarifying commercialsugar solutions. H. C. Prinsen Geerligs 88 points out that the errordue to the precipitation of laevulose only occurs in the case ofcommercial samples containing impurities precipitable by the leadreagent, whilst 0. Schrefeld 89 states that this source of error is tobe avoided by the employment of neutral lead acetate. F. Wattsand H. A. Tempany 90 recommend the Fehling-Violette solutionas the least liable to auto-reduction. I n the estimation of invert-sugar in presence of sucrose, they find that one gram of the latterexerts a reducing power equal to 0.0033 gram of invert-sugar. Thewriter and T. Rendleg1 find that sucrose has no influence on theestimation of invert-sugar until its amount exceeds 30 per cent.of the total sugars; with a mixture of equal parts of the sugar,the invert-sugar would be over-estimated by 0.8 per cent.,whilst whenthe invert-sugar only represents one per cent.of the mixture, theLast year C. J. LintnerZeitseh. ges. Brauwescn, 1907, 30, 109 ; cotnpnre A . , 1907, ii, 823.Zeitseh. Nahr. Genicssm., 1908, 16,’509 ; A . , ii, 1077.82 B i d . , 1908, 31, 53.s4 Zcitsch. ofcntl. Ckrn., 1908, 14, 150; A., ii, 543.86 Chem. B i t . , 1908, 32, 996.8G J. Amer. Chem. SOC., 1907, 29, 1744 ; A., ii, 235.s8 Intcrn. Sugar J., 1908, 10, 500 ; A . , ii, 990.gR Zcitsch. Vcr. dcut. Zzcckerind, 1908, 947 ; A . , ii, 1 O i G .91 Analyst, 1908, 33, 167 ; A ., ii, 542.J. SOC. Chem. Ind., 1908, 27, 53 ; A., ii. 236.J. SOC. Chenr. Ind., 1908, 27, 191 ; A., ii, 437ANALYTICAL CHEMISTRY. 203amount returned may be 14 per cent. too much. These resultsconfirm those of Watts and Tempany,92 who, however, dealt onlywith high proportions of sucrose to inverbsugar. The writer andG. C. Jones 93 have shown that the volumetric method of estimatingreducing sugars, using ferrous thiocyanate as indicator, gives resultsquit2 as accurats as the gravimetric method, and is far more rapid.The writer would point out, however, that with commercial products,such as molasses containing iron, the ferrous thiocyanate indicatorcannot be used. It is well known that the rotatory power oflmmlose decreases as the temperature is raised, and that at 8 7 Oit is equal but opposite in sign t o that of dextrose, so that equalamounts of dextrose and lzvulose are optically inactive a t 8 7 O .For the analysis of commercial inverbsugar, the measurement ofthe change or rotatory power with rise of temperature is useful,and the apparatus described by A.P. Sy 94 for making polarimetricdeterminations at high temperatures is much to be appreciated.T. W. Harrison and F. M. Perkin95 find that the Valenta methodis untrustworthy for the estimation of tar oils in admixture withmineral oils, the latter being not absolutely insoluble in methylsulphate.96 W. H. Emerson 97 has redetermined the solubility ofstearic acid in alcohol a t Oo for the purpose of the Hehner-Mitchellmethod.R. K. Dons98 describes a modification of his method ofdetermining the caprylic acid value in butter fat. J. Lewkowitsch 99gives constants of, and other information on, carapa oil; and1 hepublishes constants of ochoco fat from the seeds of one of theMyristicacez. M. Tsujimoto gives some constants of Japanesetea oil. Such processes of determining unsaturated fats as thoseof Hub1 and Wys (iodine absorption) are often used as mereempirical tests ; the scientific principle underlying them is broughtout, however, by S. Fokin’s experiments,3 showing that the sameresults may be obtained by determining the hydrogen value (c.c.of hydrogen a t Oo and 760 mm. absorbed by one gram of thes~bstance).~ H. Matthes and 0. Rohdich,5 working on 13 kilos.of92 LOC. eit.94 J. Amer. Chcm. Xoc., 1908, 30, 1790 ; A . , ii, 1076.95 Analyst, 1908, 33, 2 ; A . , ii, 135.9f J. Amar. Chem. Soc., 1907, 29, 1750 ; A . , ii, 236.O8 Zeitsch. Nahr. Genzcssm., 1908, 15, 75 ; A., ii, 238.93 Analyst, 1908, 33, 160 ; A., ii, 541.Compare Graeffe, Ann. Beport, 1907, 211.Analyst, 1908, 33, 184.Ibid., 313 ; A., ii, 885.Chcm. Bev. F4t FIurx. I d , 1908, 15, 224.J. nZus. I’h~y?. Chem Soc., 1908, 40, 700 ; A., ii, 637.In connexion with the chemistry of Wys’ solution, see H. Ingle, J. SOC. C‘Ima.de?.., 1908, 41, 19 ; A . , i , 199.I ? d , 1908, 27, 314204 ANNUAL REPORTS ON THE PROGEESS OF CHEMISTRY.cocoa butter, failed t o isolate any constituent to which the character-istic odour could be attributed.From the unsaponifiable matterthey isolated amyrilene, C,,H,,. H. Matthes and E. Ackermann6point out that cocoa fat contains two phytosterols, the acetyl-tetra-bromides of which melt a t 180° and 1 3 2 O respectively. Thecholesterol of butter does not form an acetyl-tetrabromide. Theethyl ester value of fats is a new constant devised by J. Raniis andL. hekl C. Fleig 8 has carriedout a series of experiments which, on the whole, support XIylius’scontention that the Camoin-Baudouin colour reaction of sesame oiland Pettenkofer’s similar reaction of bile acids are due to thefurfuraldehyde produced by the mineral acid on the sugar, but theevidence is not conclusive, since lzvulose and sucrose give betterresults than certain pentoses. He also shows that the furfural-dehyde in Villavecchia and Fabris’s reagent may be replaced byother aromatic aldehydes.C. H.Hertyg draws attention to the wide variations in theoptical activity of samples of turpentine from trees grown on thesame farm in Florida. F. W. Richardson and J. L. Bowen10 haveinvestigated various methods of detecting and estimating adul-terants in turpentine, one of the conclusions being that refracto-metric determinations of the distilled fractions give some of themost useful data.11 C. T. Bennett 12 shows that for the estimationof cineol in eucalyptus oil, Schimmel and Co.’s resorcinol methodi s untrustworthy, and he points dut that the fraction of the oilboiling between 175O and 1 8 5 O consists mainly of cineol.A. Bloch l 3estimates citral in lemon-grass oil by removing it as the bisulphitecompound. E. M. Chace14 shows that 2 per cent. of turpentinemay be detected in lemon oil by conversion into pinene nitroso-chloride and examination of the crystals mounted in olive oil underthe microscope. P. Jeancard and C. Satie15 give the opticalactivity, specific gravity, and ester value of samples of Alpinelavender oil distilled in different years, whilst M. Daufresne 16 hasdetermined some constants of French and German oil of tarragon.for the detection of cocoanut oil.ti I!cr., 1908, 41, 2000 ; A . , i, 637.Zcitsch. Nahr. GenzLssm., 19OS, 15, 5 i 7 ; d., ii, 641.8 Bull. SOC. chim., 1908, [iv], 3, 984, 992 ; A . , ii, 994.‘J J. Amer. Chcm. Soc., 1908, 30, 8 6 3 ; A., i, 434.lo J.SOC. Chem. Ind., 1908, 27, 613. ~l2 Chemist and Drisggist, 1908, 72, 55.l 3 IIulZ. Sci. Pharrnacol., 1908, 15, 72 ; A . , ii, 782.14 J. A m e T . Chem. Xoc., 1908, 30, 1475 ; A., ii, 908.l5 Bull. J‘oc. chim., 1908, [iv], 3, 155 ; A . , ii, 232.l6 Bid., 300 ; compare d., i, 436.See also A. K. Turner, Oil mid Coloirr Trades J . , 1905, 503ANALYTICAL CHEMISTRY. 205S. S. Pickles17 gives some constants of origanum oil, from whichhe has isolated a new terpene, origanene.Among papers dealing with the chemistry of the proteins, thefollowing may be mentioned: G. T. Matthaiopoulos 18 makes use ofthe fact that casein forms a definite compound with sodiumhydroxide, in order to estimate it (volumetrically) in milk. Hisresults confirm Laqueur and Sackur’s observation that the molecularweight of casein is 1135.6. F. Cross, E. J. Bevan, and J. F.Brigs19 have extended the work of F. Raschig,2O and have shownthat, like ammonia, the proteins and their derivatives form chloro-amines by the action of hypochlorites. These chloroamines liberateiodine from potassium iodide, and the presence of proteins in anorganic tissue may be located by first steeping it in an acid solutionof bleaching powder, and, after washing, removing the excess ofchlorine by immersion in 2 per cent. sodium phosphate solutiona t 4 5 O , and finally treating it with a solution of potassium iodideand starch: a blue stain will be produced in those parts in whichproteins occur. R. H. Aders Ylimmer and F.H. Scott21 find thatwhen phosphoproteiiis are digested with a one per cent. solutionof sodium hydroxide for twenty-four hours a t 3 7 O , the whole ofthe phosphorus is eliminated as phosphoric acid. This serves,therefore, as a means of distinguishing these compounds from thenucleoproteins. S. P. L. Sorensen 22 proposes to measure the amino-acid formed by the hydrolysis of proteins by a method based onH. Schiff’s observation that the basic function of these compoundsis annulled by the formation of the group N:CH, after treatmentwith formaldehyde, the carboxyl group being then titrated withalkali hydroxide, using phenolphthalein or thymolphthalein asindicator. The reaction of the amino-acids, formaldehyde, andalkali is reversible, but when thymolphthalein is employed asindicator an end point (bright blue colour, obtained by four dropsof rV/5 barium hydroxide in excess of that required to producethe first colour indication) may be chosen such that the concentra-tion of the hydrogen ions is as low as 10-9.7.23 To measure therate of proteolysis, Griitzner has proposed to estimate the carminliberated from fibrin stained with that dye, but H.E. Roaf24 statesthat Congo-red may with advantage be substituted for carmin.I’ P ~ v c . , 1903, 24, 05.Zeitsch. and. G‘lum., 1908, 47, 492 ; A., ii, 783.J. SOC. CIienr. I d . 1908, 27, 260 ; A . , i, 374.211 L‘T., 1907, 40, 4586 ; Citein. Zeit., 1907, 31, 126 ; Zeitszh. mayew. Chem.,’’? Big cJwm. Z E Z ~ ~ L . , 19Oi, 7, 45 ; A . , i, 115.“3 See, fuitlier, S.P. L. Sorensen aiid H. Jesseri-Hansen, ibid., 1908, 7, 407 ; A . ,1907, 20, 2065 ; A., ii, 30. 21 Trans., 1908, 93, 1699.ii, 334. 24 Bioclzc?~~. J., 1908, 3, 188 ; A., ii, 743206 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.I n the analysis of organic substances, increasing use is beingmade of physical methods, and numerous papers. have appearedrecently on the estimation of total solids in molasses, wort, beer,wine, spirits, etc., by means of the refractometer, which can alsobe used for the estimation of alcohol. The method is principallyto be recommended on the score of its rapidity, although in thehands of a skilful worker in a properly equipped laboratory thedetermination of specific gravity can be carried out very rapidly.On account of the heterogeneous composition of many commercialproducts, notably the varying amounts of mineral matters present,special tables must be constructed for each kind of product.Forpractical purposes the assumption that all sugars in solution oflike concentration have the same refractive index is quite justified.Tables have been prepared for sugar products by H. Main25 andby H. C. Prinsen Geerligs.26 H. Bryan,27 using Geerligs’s tables forhoney, syrups, and molasses, finds that the results are in closeraccord with the actual determinations of solid matter than arethe values obtained by the specific gravity method from Brix’stables. It should be pointed out, however, that this is not due toany defect in the specific gravity method.C. Mai and S. Rothen-fusser28 give a valuable rQsum6 of the application of the refracto-metric method of detecting the addition of water to milk.From the examination of 5000 samples, they find that thenormal refractometer value for milk is 39 divs.; any samplegiving a lower reading than 36.5 divs. contains added water.The process fails with sour milks. The application of themethod to the estimation of extract in wort is dealt with by0. M0hr,2~ and to that of alcohol and extract in beer and spiritsby J. Race.30 A. Frank31 observes that the indirect estimation ofalcohol by refraction before and after expelling the alcohol isuntrustworthy. M. Duboux and P. Dutoit 32 describe a new methodof estimating alcohol in wine; it depends on the determination ofthe temperature a t which a clear solution is formed when a fixedamount of the wine distillate is added to a fixed amount of amixture of aniline or nitrobenzene and alcohol.Some useful datain the pyknometric estimation of alcohol in fermented liquids, witha description of new apparatus, is given by W. Antoni.3325 International Sugar J., 1907, 9, 481.27 J. Amer. Chena. Xoc., 1908, 30, 1443.Zeitsch. Nahr. Cenzusm., 1908, 16, 7.Wochensch. Brau., 1908, 25, 454.30 J. SOC. Chm. Jnd., 1908, 27, 544, 547 ; A . , ii, 738.Chem. Zeit., 1908, 32, 569; A., ii, 687.32 Ann. Chim. anal., 1908, 13, 4 ; A,, ii, 136.:ja J. Amr. Chem. Soc., 1908, 30, 1276; A . , ii, 902.26 Ibid., 1908, 10, 68ANALYTICAL CHEMISTRY. 207M. E. Pozzi-Escot 34 describes a volumetric method of estimatingtart8aric acid in wine, whilst the Goldenberg method has beeninvestigated by the Cheniische Fabrik vorm.Goldenberg, Geromontand C0.,35 and conditions laid down whereby accurate results maybe obtained. L. Gowing-Scopes 3G has submitted J. von Ferentzy’smethod of estimating tartaric acid in presence of other acids37 toa critical examination, and finds it trustworthy. A. Hubert 38 hasmade the interesting observation that citric acid occurs naturallyin wines. This has been confirmed by H. Astruc 39 and by G.Denigks,*O who suggests that the reason the acid cannot be detectedin old wines is to be ascribed to bacterial influences. E. Dupont 41states that Denigks’s mercury method of detecting citric acid maybe made approximately quantitative. G.Favrel 42 proposes a testwhich depends on the forrnatisn of acetonedicarboxylic acid (whencitric acid is heated with sulphuric acid). This ketonic acid gives aviolet coloration with ferric chloride.P. Dutoit and I$. Duboux43 show that by adding successivequantities of barium hydroxide to a wine, and determining its elec-trical conductivity after each addition, it is possible to estimateconsecutively the sulphates, total acidity, and tannin substancespresent. Under the name of “ abrastol,” calcium P-naphthol-a-sulphonate is added to wine as an antiseptic and precipitant oftartrates. The acid may be extracted by amyl alcohol, andidentified by the reddish-violet coloration formed on evaporatingwith mercurous nitrate.I n connexion with brewing materials, A.C. Chapman44 hasdevised a valuable method of estimating tannin in hops byweighing the cinchonine compound, and G. C. Jones4“ has com-municated two papers on malt analysis.The paper by J. S. Ford and J. M. Guthrie,46 on the biochemistryof barley, adds considerably to our knowledge of the nature of theamylolytic enzyme of barley, and the methods they adopt are mostinteresting and suggestive. J. Wohlgemuth’s ~uggestion,~~ tomeasure diastatic activity by estimating the quantity of a diastatic. -34 32111. Xoc. chim. Belg., 1908, 22, 218 ; A., ii, 740.35 Zeikch. anal. Chem., 1908, 47, 57 ; A., ii, 237.36 Analyst, 1908, 33, 315 ; A.! ii, 905.:;’ See Ann. Report, 1907, 217.39 Ibid., 224 ; A., ii, 640.41 Ibid., 338 ; A., ii, 904.43 Compt.rend., 1908, 147, 134 ; A., ii, 781.44 J. Inst. Brewing, 1907, 13, 646.45 Ibid., 1908, 14, 9, 13.46 ]bid., 61 ; A!., ii, 218.Ann. Chiin. anal., 1908, 13, 139 ; A., ii, 544.4o Ibid., 236 ; A., ii, 640.4iz Ibid., 177 ; A . , ii, 640.B o c h e n t . Zeitsch,., 1908, 9, 1 ; A.: ii, 444208 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.solution necessary to hydrolyse soluble starch to the stage a t whichthe products no longer give a coloration with iodine, assumes thatthis is a linear function of the time, which has not been proved,and the same argument applies to W. A. Johnson’s method,lBalthough this author does compare the iodine results with thoseobtained by the cupric reduction method. C. J. Lintnef49 showsthat the law of proportionality for malt diastase holds good up toa reducing power of 35 (calculated as maltose). He adopts thewriter’s method of estimating diastatic activity. J. L. Baker andH. F. E. Hulton50 and J. S. Ford and J. M. Guthrie51 havepublished two most important papers on the question of theso-called “ strength ” of flour in its relation to enzymes. L. Briegerand J. Trebing52 call attention to the antitriptic power of humanblood-serum, and von Bergmann and K. Meyer 53 give a methodfor the measurement of its activity.W. Thorner54 describes a method of estimating water in foods,consisting in distillation with petroleum and collection of the dis-tillate in a graduated receiver, in which the water separates and itsvolume may be read off. The method is similar to that ofHoffrnann,55 and of Aschnian and Arend.56H. D. Richmond,57 in his annual report on milk, based this yearon the analysis of 35,331 samples, adds to our knowledge of thesubject. EIe has examined eight samples of human milk, and pointsout that the fat varied from 1.7 to 5.7 per cent. He draws attentionto a preservative xhich he has found in certain samples of milk,naiiiely, formic acid, mixed with dextrose. W. R. G. Atkins 5sstates that determinations of the freezing point and specific gravityof milk are sufficient to show if water has been added or Patremoved. The freezing point of milk is practically a constant,namely, 0.55O.A. Kreutz59 shows that after heating cocoa in a, flask on thewater-bath with chloral alcoholate until a homogeneous paste isobtained, the fat may be extracted by warming with successivequantities of ether. The chloral is all driven off by heating the4g J. Amcr. Chem Soc., 1908, 30, 798 ; A!., ii, 743.ao J. s’oc. Chcm I n d . , 1908, 27, 368.5l Ibid., 389.jy Bcrliz Ktin. TVocheiwdL., 1908, 45, 1041.B3 Ibid., 1673.ri4 ZciilscJt. nngctc. Chcwz., 1908, 21, 148 ; A . , ii, 222.35 Wochemch. Brau., 1902, 19, 572.jti Ufiem. Zeit., 1906, 30, 953 ; A . , 1906, ii, 814.57 A?zuEyst, 1908, 33, 113.s8 Chenb. News, 1908, 97, 241 ; A . , ii, 641.ZeztsclL. yes. Braz~wesen, 1908, 31, 421.Zeitseh. Nahr. Geitussna., 1908, 15, 680 ; 8., ii, 641ANALYTICAL CHEMlSTRY. 209fat at 1 1 0 O . The results agree well with those obtained by theordinary method. Later,6" he shows that the theobromine isextracted along with the fat by this method; when treated withcarbon tetrachloride the fat alone is dissolved, leaving the theo-bromine, which may be weighed.P. Biginelli61 draws attention to the fact that when tannic acidis used as a precipitant for alkaloids in toxicological investigations,it also forms insoluble compounds with certain solvents and mineralacids. L. Dreyer 62 describes the differentiation of tubercular fromordinary pus by its behaviour towards Millon's reagent. It is wellknown that the formation of indole from proteins is a characteristicof certain microbes, and C. Porcher63 proposes to detect indole inpus by its colour reaction with p-dimethylaminobenzaldehyde.ARTHUR R. LING.6o Zcitsch. Nahr. Gcnttssm., 1908, 16, 579.G1 Gazxetta, 1907, 37, ii, 506 ; A., i, 40." illilnch. Mcd. TVocke,wcl~., 1906, 55, i 2 8 .63 C'ompt. rend., 1908, 147, 214 ; A., ii, 769.REP.-VOL. V.