摘要:

ANNUAL REPORTSax THEPROGRESS O F CHEMISTRYANNUAL REPORTSE:. C. L’. BALY, Y.R.S.H. &I. DAWSON, D.Sc., 1% I).F. G. HOPKINR, M A . , M.E., D Sc.,J. C. IRVINE, D.Sc., P1i.D.F. R. s.ON THEPROGRESS OF CHEMISTRYF O R 1916.ISSUED BY THE CHEMICAL SOCIETY.N. H. J. M I L L E R , Pli D.C. A. MITCHELL, F.I.C.F. L. PYYAN, D.Sc., Ph.L).P. SODDY, M A , F.R.S.A. W. STEWART, D.Pcclfommiftre o f @ublitatiau :A. C’HAMTON CHAPMAN.A. W. CKOSSLEY, D.Sc., P11.D., F.R.S.F. G. DOSNAN, M.A., PIi.D., F.R.S.BERNARD DYER, D.Sc.hl. 0. FORSTEP., I>.Sc., Ph.D., F.R.S.A. HARDEN, D.Sc., Pti.D., F. I.I.S.T. 31. LOWRY, D.Sc., F.R.S.J. C. PHILIP, D.Sc., Ph.D.F. L. PYMAN, D.Sc., Ph.D.A. SCOTT, M.A., D.Sc., F.R.S.G. SENTER, D.Sc., Ph.D.S. SMILES, D.Sc.J.F. THOKPR, D.Sc., Ph.D., F. . P .@hifor :J . C. CAIS, D.Sc., Ph.D.Sih-6;bitor :A . J. GREENAWAY.&3sistrrnt Sub-fibitor :CLARENCE SMITH, D.dc.Vol. XIII.LONDON:GURNEY t!k JACKSON, 33, PATERSOfilTER ROW, E.C.1917PRINTED I N QRFAT BRITAIN BXRICHARD C L I ~ AXD SONS, I,IMITED,aRUkSWTCK STREET, STAMFORD STRFET, 8 . C .AND RVNO4T. YI'FFOI h CONTENTS.PAGEGENERAL AND PHYSIC’AL CHEMISTRY. By H. hl. DAWSOX, n.Sc.,P h . D . . . . . . . . . . . . . 1INORGANIC CHEMISTRY. By E. C. C. BALY, F.R.S. . . . . 31Part ~.-ALIPHATIC DIVISION. By J. C. IRVINF,, D.Sc.. Ph.D. . . 66Part II.-HOMOCYCLIC DIVISIO?;. By F. L. PYMAN, D.Sc., Ph.D. . . 97Part III.-HETEROCYCLIC DIVISION. By A. W. STEWART, D Sc. . . 129ANALYTICAL CHEMISTRY.By C. A. MITCHELL F.I.C. . . . . 16.1PHYSIOLOGICAL CHEMISTRY.F.R.S. . . , . . . . . . . . . 195ORGANIC CHEMISTRY :-By F. G. HOPKINS, M.A., M.R., U.Hc.,AGRICULTURAL CHEMISTRY AND VEGETABLE PHYSIOLOGY.By N. H. J. MILLER, P1i.D. . . . . . . , . 219RADIOACTIVITY. By Biier>~ixc~ SODDY, M.A., F.K.S. . . . . 24TABLE OF ABBREVIATIONS EMPLOYED I N THEA B B 11 E VI .4T ED TITLE.A . . . .Antcr. J. Physiol.Amer. J. Sci. .Analyst . .knnnlen , .Ann. Bot. . .Ann. Chim. .Ann. Chiin. nnnl.Ann. Chiin. ApplicntnAnn. Chim. . .Ann. Inst. Pnsteicr .Ann Physik . .Ann. Physique . .Ann. Report . .Arch. Int. Med.Arch. Pharm. . .Arkiv. Kein. Jf in. Geol.Atti I?. Accnd. Liiicei .Ber. . , . .Ber. Deut. hot. Ges. .Bcr. Deut.piiarm. Qes.Ber. Deut. yhysika7. Ges.Bied. Zentr. . .Ap0th:Zeit. . .Bicchem. Bwll. . .Biochcm. J. . .Biochem. Zeitsch. .Bull. Soc. chim.Centr. Bakt. Par. ,Chenb. News . .Chem. Weekblnd .Chm. Zeil. . .Chem. Zentr. . .Compt. rend. . .Gcmzetta . . .Gummi- Zeit. . .Inter. Zeitsch. Phys. -chem.Biol. . . . .Jahrb. Rndioaktiv. Elek-tronik . . . .J. Anzer. Chem. SOC. . , . J. Hiol. Chein. . . .J. B d . Agric. . . .J. Agric. S’ci. . .REFERENCES.JOURNAL.Abstiacts in Journal of the Chemical Society.*American Journal of Physiology.American Journal of Science.The Analyst.Justus Liebig’s Annalen der Chcmie.Annals of Botany.Annales de Chiniie.Annalrs de Chiiiiie analytique appliqnee a l’Industiie,Annali di Chimica Applicata.Aiinales dc Chimie.Annalcs de 1’Institut Pasteur.Annalen der Physik.Annales de Physique.Annual Reports of the Chemical Society.Apotheker Zeitung.Archives of Internal Medicine.Archiv der Phnrmazie.Arkiv for Xemi.Mine1 alooi och Geologi.Atti della Reale Arcade& dei Lincei.Berichte der Deutschen chemischen GesellschaftBerichte der Dentschen botanischen Gesellschaft.Berichte der Deutsclien pharmazeutischen Gesell-Berichte dcr Deutsclien physikalischen Gcsellschaft.Biedermann’s Zentralblntt fur Agrikulturchemie undBiochemical Bulletin.The Biochemical Journal.Biochemische Zeitwhrift.Bulletin de la Soci6t6 chimique de France.Centralblatt fur Bakteriologie, Parasitenkunde undChemical News.Cheiniscli Weekblad.Chemiker Zeitung.Chemisches Zentralblatt.Comptes rendus hebdomadaires des Seances deGazzctta chimica italiana.Gummi-Zeitung.Internationale Zeitschrift fur physikalisch-chemischeJahrbuch der Radioaktivitat und Elektronik.Journal of Agricultural Science.Journal of the American Chemical Society.Journal of Biological Chemistry, New Pork.Journal of the Board of Agriculture.a I’Agriculture, B la Pharniacie et la Hiologie.schaft.rationellen Landwirtschafts-Betrieb.Infektionskrankheiten.I’Acadkrnie des Sciences.BiologieThe year is not inserted in references t o 19ltiviii TABI,E OF ABBREVIATIONS EMPLOYED IN THE REFERENCES.ABBREVIATED TITLE.J.Ind. Eng Chcm. . .J. PIiccrm. Chine. .J. Phgsicnl Chem. . ,J. Physiol. . . .J. p r .Chcnz. . . .J. &y. Agric. SOC. . .J. Buss. Phys. Chesn. Soc. .J. SOC. Chcm. Ind. . .J . IVcicshington Acnd. Sci. .Kollo id- Ze; itsch. . .Lnndiu. Versuchs-Slat. .Monatsh. . . . .dlon. Xci. . . . .Ntcovo Cisn. . . .Pharm. J. . . . .Phnrm. TVceXblnd . .Phil. Mug. . . .Physikccl. Zeitsclz. . .P . . . . . .Proc. Casnb. Phzl. SOC.Proc. K. Akad. Wctcnsch.Asnstcrclasn. . . .Proc. Nut. Acnd. Sei. . .Proc. Roy. Soc. . . .Aec. trav. chiin. . . . Proc. Roy. SOC. Edit?.S c i . Proc. Boy. Dobl. SOC. .Sitsicn ysber. K. Aknd. 1Yiss.Berlin. . . . .Soil. Sci. . . . .Staz. spcr. agrnr. Itnl. .T . . . . . .Trans. Faraday SOC. . .Trans. Roy. Xoc. Cn7cndn .Zcitsch. anal. Chcm. . .Zeitsch. nngcw. Chcna .Zeitsch. nnorg.Chent,. . .Zeitsch. BioZ. . . .Zeitsch. Eleklrochenz. . .Zeitsch. Kryst. Min. . .Z4tscl~. physiknl. Chcm. .Zeitsch. physiol. Chem .Zeilsch. w iss. M? kroskop. ~JOURNAL.Joiirnal of Indnstrial and Enginerring Chemistry.Journal de I’harniacie et de Chiniie.Journal of Physical Chemistry.Journal of Pliysiology.Journal fiir praktischc Cliernie.Jorirnal of the Royal Agricul!mal Society.Journal of the Physical and Chemical Society ofJournal of the Society of Chemical Indust,ry.Journal of the Washington Academy of Sciences.Kolloid-Zt~itschrift.Die laiidwirtschaftlichen Vcrsuchs-Stationen.Monatshefte fiir Chemie und vcrwandte Theile aiidercrNonitrnr scientifiqnc dc Qiirsneville.I1 Ruoro Cimento.The Pharmaceutical Jonnial.Phnrmacl ut isch \Vcrkblacl.l’hilosophical Magazine (The London, Edinburgh andPh ysikalische Zei t schrift.Proceedings of the Cheniical Society.Proceedings of the Cambridge Pli ilosophical Society.Koninklijke Akadeniie van Wetenschappen te Anistcr-Proccedings of the National Acadt my of Sciences,Procecdings of the Royal Society.Proceedings of the Royal Society of Edinbnrgli.Raueil des travaus chiiniqnes des Pays-Tlas et de laScientific Proceedings of the Royal Dublin Society.Sitzungsbericlite der Koniglich Preussischen AkademieSoil Science.Stazioiii sperimentali agrarie Italiani.Transactions of the Chemical Society.Transactions of the Faraday Society.Transactions of the Royal Society of Canada.Zcitschrift fiir analytische Chemie.Zcitschrift fiir angewaiiclte C’hemie.Zeitschrift fiir nnorganische Chemic.Zeitsclirift fiir Biologic.Zeitsclirift fiir Elektrochemie.Zeitschrift fiir Iirystallographic und Mineralogie.Zeitsclirift fiir physikaiische Chcmie, StochiometrieHoppe- Seylcr’s Zeitschrift fiir physiologische Chemie.Zcitschrift fiir ~~,isseiischaftliche EIikroekuyie u u dKussia.Wissensch aften.D II 1 ili n ) .dam.Proceedings (Engli+li version).IVasliington.Helgiq tic.der Wissenschaften zii Berlin.iind Verwan~itsch~ft.;lehre.niikroslropische Technik

ISSN:0365-6217    

DOI:10.1039/AR91613FP001

出版商:RSC    

年代:1916

数据来源: RSC

摘要:

INORGANIC CHEMISTRY.ALTHOUGH the number of papers dealing with inorganic chemistrythat have been published during this last year shows a markeddecrease, many of them deal with subjects of great interest. Not-ably is this the case with the investigations of the atomic weightsof the isotopes of lead and thorium. The last doubt must now bebanished from the mind of the most pronounced sceptic as to thedefinite existence of isotopes with identical chemical properties,but with markedly different atomic weights. The discovery re-corded during this year of the fact that common lead and radio-lead have the same atomic volume would seem to give the finalproof.The enunciation of this theory by Soddy and by Fajans, and itsproof by Richards and by Honigschmid, must surely rank as oneof the most striking advances that has taken place in chemistryduring recent years.Moreover, its intrinsic value in supportingthe modern views as t o the structure of atoms is obvious.In the following Report the opportunity has been taken ofincluding a review of the work which has been carried out duringthe last four years on the chemistry of the rare earths.Atomic Weights.The International Committee, in their report for 1916: onlymake one change in the list of atomic weights as printed for 1915.They recommend that the value of 93.1 be adopted for the atomicweight of columbium. This change is based on work described inlast year’s Report.:! From the determinations of atomic weightsthat have been made in the past twelve’ months the following maybe selected.Nydrogen.-A very accurate determination has been made ofthe ratio of combining volumes of hydrogen and oxygen.3 TheT., 1916, 109, 777.E.F. Smith and W. K. van Haagen, J. Arner. Chem. Soc., 1915, 37,1783 ; A . , 1915, ii, 692.F. P. Burt and E. C. Edgar, Pkil. Trans., 1916, [A], 216, 393 ; A . , ii, 427.332 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.hydrogen was prepared by the electrolysis of a solution of bariumhydroxide which had four times been recrystallised. The gas waspurified either by means of cocoa-nut charcoal a t the temperatureof liquid air or by passage through an electrically heated palladiumtube. The oxygen was prepared by the electrolysis of bariumhydroxide or by heating potassium permanganate. It was puri-fied by condensation and subsequent fractionation.The volumesof the gases were actually measure'd a t Oo and 760 mm., in ordert o obviate the necessity of correction f o r temperature and pressure.One volume of oxygen was exploded with rather more than twovolumes of hydrogen, and these volumes were measured consecu-tively in the same apparatus, and a t the end the volume ofresidual hydrogen was measured I n this way the formation ofozone, hydrogen peroxide, and oxides of mercury was eliminated.From five series of experiments the volume ratio was found to be2.00288 a t Oo and 760 mm. From Morley's values for the weightsof 1 litre of hydrogen (0.089873) and oxygen (1.42900) this ratiogives the atomic weight of hydrogen as 1.00772. On the otherhand, recent work renders probable a somewhat higher value foroxygen, 1.42905, this being the value proposed by Germann as aslightly weighted mean of his results combined with those1 ofRayleigh and of Morley.4 If this figure is adopted, the atomicweight of hydrogen is found to be 1.00769, and this would seemto be the moat accurate value yet obtained.It should be pointedout t h a t a sixth series of experiments was carried out with oxygenin slight excess, and the result obtained from these shows t h a tthere is no constant error introduced by the us0 of the slightexcess of hydrogen.I n this connexion reference may be made t o the pressure exertedby the same volume of gas in vessels of different shape. Morleyfound thO ratio of combining volumes of hydrogen and oxygen tobe 1 : 2.00023, while Scott found the ratio to be 1 : 2.00285.Morleymeasured the gases in a eudiometer t u b , whilst the latter usedspherical vssels, and Morley suggested t h a t the difference mightin some way be due to the difference in shape of the measuringvessels. A careful comparison has been made of the volumeoccupied by a gas in a system of tubes with t h a t which it occupiesin a bulb.5 The difference does not exceed 1 in 10,000, whilst theabove ratios differ by 1 in 1000.Brorniiw-A series of determinations has been made of thedensity of hydrogen bromide.6 The gas was prepared by theaction-4. F. 0. Germann, J . Chim. phys., 1914, 12, 6 6 ; A., 1914, ii, 454W. A.Noyes and L. C. Johnson, J . Amer. Chem. SOC., 1916, 38, 1017;E. Moles, Compt. rend., 1916, 162, 686; 163, 94; A., ii, 314, 526.A., ii, 3751NORGANIC CHEMISTRY. 33of water on phosphorus tribromide, bromine on hydrogen sulphide,bromine on naphthalene a t the ordinary temperature, and onparaffin heated a t 200O. The gas was very carefully purified ineach case. As a mean of thirty-three determinations the weightof 1 litre of Oo and 760 mm. was found to be 3~64442~0~00013grams. The weight of the gas under diminished pressures was alsodetermined and the coefficient of compressibility obtained. Themolecular weight of the gas was thus found to be 80.934, whencethe atomic weight of bromine is found to be 79.926 (H= 1.008).Cadmium.-In last year's Report reference was made to thevalue of 112.417 obtained for the atomic weight of cadmium byBaxter and Hartmann,7 and to the criticism made by Hulett andQuinn, who obtained the figure of 112-3.8 During this year theformer authors give the results of some further measurements bythe same method, namely, the electrolysis of cadmium bromidesolutions with a mercury cathode.As a mean of twelve determina-tions, the atomic weight was found to be 112'407.9 An analogous ,series of determinations with cadmium chloride gave the figure112.413. On the other hand, a low result for the atomic weighthas been published by (Echsner de Coninck and GQrard.10 Themethod adoDted was to dissolve the cadmium in sulphuric acid, theresulting solution being treated with hydrogen sulphide.Thecadmium sulphide, after washing, was dissolved in concentratedhydrochloric acid, the excess acid evaporated, and the cadmiumprecipitated as carbonate by the addition of a large excess ofammonium carbonate. A weighed quantity of the pure cadmiumcarbonate was reduced to metal in a current of pure hydrogen.As a mean of five determinations, the atomic weight was found t obe 112.32. It should be pointed out, however, that owing to theamounts employed a small error in the weighing would affect theresult to an abnormally great extent, and therefore perhaps toogreat value should not be placed on this result.Zim-By a method similar t o that described under cadmium,namely, the electrolysis of a solution of the bromide, the atomicweight of zinc as a mean of eight determinations was found t obe 65*388.11 The zinc befors use was very carefully purified, andG.P. Baxter and M. L. Hartmann, J. Amer. Chem. Soc.. 1915, 37, 113 ;A., 1915, 11, 98.a G. A. Hulett and E. L. Quinn, ibid., 1915, 37, 1997 ; A., 1915, ii, 771.G. P. Baxter, M. R. Grose, and M. L. Hsrtrnann, ibid., 1916, 38, 867;lo W. mchsner de Coninck and GBrard, Compt. rend., 1915, 161, 676;G. P. Baxter and M. R. Grose, J . Amer. Chent. SOC., 1916, 38, 868 ;A., ii, 327.A., ii, 33.A., ii, 327.HE P.-VOL. XIII. 34 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.a spectroscopic test showed the presence of only minute traces afcadmium, estimated a t 0.001 per cent.Uranium.-The previous determinations of the atomic weight ofthis element have been made with material prepared from pitch-blende.It has been thought worth while to repeat the determina-tions with uranium obtained from another source, and the crystal-lised uranium ore from Morogoro, referred to in last year's Report,was used. The Pb : U ratio would lead to the conclusion that thisore is eight hundred million years old as against two hundred andfifty million years for pitchblende. This Morogoro ore is almostfree from impurities. The values obtained from two series ofdeterminations were 238.043 50.018 and 238.159 +0.023,12 themethod being the determination of the ratio UBr, : AgBr, as previ-ously described. The value obtained in the second series is veryclose to the old value of 238.175,13 which led the InternationalCommittee to adopt the figure 238.2.A very interesting series of papers has appeared on the atomicweights of the isotopes of thorium and lead, which may be describedin some detail. The case of thorium and its isotope ionium mayfirst be dealt with.Measurements have been made of the atomic weight of thoriumobtained both from minerals rich in uranium and from mineralspoor in uranium.14 The former specimens of thorium are rich inionium.Now thorium and ionium are isotopes and identical intheir chemical and physical properties, but they should havedifferent atomic weights. The method employed was to obtain theratios ThBr, : 4Ag and ThBr, : 4AgBr, the purification of the mate-rial being similar to that adopted for uranium bromide, andalready described.The values obtained f o r the atomic weightwere as follows:ThBr, : 4Ag, 232.152.ThBr, : 4AgBr, 232.150.(Th*Io)Br4 : 4Ag, 231.507.(Th*Io)Br, : 4AgBr, 231'502.It is clear from these figures that ionium has a lower atomicweight than thorium. The mean value for the atomic weight ofthorium, 232.151 *0-0165, is considerably lower than the inter-national value of 232.4, but it is believed to be more correct. Thedifference between the two values is discussed a t some length in a12 0. Honigschmid and (MlIe.) S . Horovitz, Monatsh., 1916, 37, 185 ; A.,0. Honigschmid, Zeitsch. EZektrochem., 19 14, 20, 452 ; Compt. rend. ,ii, 484.1914,158,2004 ; A., 1914, ii, 662.l4 Ibid., 1916, 22, 18; A , , ii, 40iINORGANIC CHEMISTRY.35later paper.15 m e international value depends on the conversionof anhydrous thorium sulphate into the oxide, and this methodis adversely criticised. The determinations with thorium bromidewere repeated with material from repeatedly recrystallised thoriumammonium nitrate and with a preparation purified by the sulphatemethod, followed by the iodate method. Both gave identicalresults, Two independent series, one of twelve and the other offifteen experiments, gave 232.15 _+ 0.01 6 and 232.12 * 0.014, thelatter value being the more probable. As regards the atomicweight of ionium, the question is discussed in a third paper. Theatomic weight should be edther 238.18 - 8 = 230.18 if calculatedfr,om uranium, or 226.0 + 4 = 230.0 if calculated from radium.Theatomic weight was again determined with an ionium preparationmade by Auer von Welsbach, which was spectroscopically identicalwith the thorium used in the previous work. The atomic weight ofthis material was found to be 231-51+0-014.16 I f 230 were thetrue atomic weight of ionium, this would mean 30 per cent. ofionium and 70 per cent. of thorium in the preparation.Lead.-A further investigation has been made of the atomicweight of lead from radioactive minerals.17 Four samples of radio-lead were obtained from widely differing sources, namely, Austra-lian carnotite, American carnotite, Norwegian cleveite, and Nor-wegian broggerite. Two investigations were mads with commonlead from galena as a control, ahd the values obtained for theatomic weight were 207.179 and 207.188, giving a mean of 207.183.The results obtained with the four samples of radio-lead may betabulated as follows :Australian carnotite (four analyses), 206.342.American carnotite (two analyses), 207.004.Norwegian broggerite (one analysis), 206-122.Norwegian cleveite (two analyses), 206-084.The value obtained for the lead from cleveite is essentially the sameas that obtained by Honigschmid and Horovitz 18 with radio-leadfrom broggeAte, namely, 206.06, whilst the Norwegian broggeritehas given a value so near that it may be presumed that the lead isof the same type.The two carnotites, on the other hand, givewidely differing results. The Australian carnotite certainly con-tained galena, and therefore the radio-lead may have been mixedHonigschmid and (Mlle.) S.Horovitz, Monatsh., 1916, 37, 305; A.,ii, 610.I6 Ibid., 335 ; A., ii, 610.l7 T. W. Richards and C. Wadsworth 3rd, J . Amer. Chem. Soc., 1916, 38,0. Honigsohmid and (Mlle.) S. Horovitz, Monatah., 1916, 36, 355 ; A , ,2613.1915, ii, 635.c 36 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.with coamon lead. The value of 206.34 would be given by a mix-ture of one part of common lead with three parts of radio-lead,such as is obtained from the cleveite. On the other hand, theAmerican carnotite is very puzzling, since the value obtained wouldpoint to a mixture of one part of radio-lead with five or six partsof common lead. This would mean a very large amount of galenain the carnotite.Some uncertainty about the source renders itimpossible to say whether this explanation is satisfactory. Alter-natively, it might be assumed that there is a third variety of radio-lead, with yet higher atomic weight, such as that indicated bySoddy and Hyman.10 Measurements were made of the radio-activity of the four samples of radio-lead in the form of chloride,and it was found that the radioactivity of the lead from theAmerican carnotite was the greatest and that from the Australiancarnotite the least. The radioactivity cannot, therefore, be duet o the isotope of lowest atomic weight. It is more probably due toradium-E. Very careful investigation showed that there is nodifference whatever in the spark spectra of ordinary and radio-lead.Concurrently with the atomic weight determinations of radio-lead, measurements have been made of its density.20 It is foundthat the density of radio-lead from Australian carnotite as deter-mined by the pyknometer method a t 19.94O is 11.288, whilst thatof ordinary lead is 11.337.The1 density of the radio-lead fromcleveite is 11.273. From the values of the atomic weights givenabove, the atomic volumes of common lead and the radio-leadsfrom carnotite and cleveite are 18.277, 18-279, and 18.281 respec-tively. I n other words, the atomic volumes of lead and its isotopeare identical. This fact is highly significant in view of the disin-tegration theory put forward by Soddy and by Fajans.A very important series of papers has been published on theerrors which affect atomic weight determinations.21) 28,23, 24 I n thefirst of these the various errom are discussed which are involvedin weighing and the reduction of the weights to a vacuum.It isshown that there are ten sources of error, and the probable valuesof these are given. I f these errors are not eliminated, an error of0-19 mg. is possible in any given determination. It is pointed outF. Soddy and H. Hyman, T., 1914, 105, 1402.2o T. W. Richards and C. Wadsmorth 3rd, J. Amer. Chem. SOC., 1916,38,221,1658; A., ii, 251, 566.P. A. Guye, J . Chim. phys., 1916, 14, 2 5 ; A., ii, 385.22 T. RBnard and P. A. Guye, ibid., 1916, 14, 55 ; A., ii, 386.23 P. A. Guye, ibid., 1916, 14, 83 ; A., ii, 386.24 P. A. Guye and F.E. E. Germann, ibid., 1916, 14, 195, 204; A., ii, 445,432INORGANIC CHEMISTRY. 37that values of atomic weights purporting to be accurate t o 1 partin 500,000 are somewhat illusory.It is furt1ie.r pointed out that recent determinations of theatomic weight of silver are not sufficiently concordant, and it issuggested that this is partly due to the presence of gaseous impuri-ties in the metal. Experiment shows that after bubbling hydrogenthrough molten silver, certain gases, notably carbon monoxide, areretained in sufficient amount to account for appreciable variationsin experimentally determined atomic weights. Further, the gascontent of the silver varies throughout the cooled mass. Consider-able interest attaches to the question of purity of substances, andthe classic work of de Gramont on the ultimate spectrum lines hasenabled the use of the spectroscope as a means of detecting impuri-ties, since it is now known which spectrum lines show themselveswhen impurity is present in very small amounts.de Gramont givesthe impurities which are present in forty-two solid elements, divid-ing them into two classes, namely, ‘those which are generally pre-sent and those which are always present whether the element isfree or in the form of a simple salt, however rigid the purificationhas bee11.~5 He directs special attention to the case of silver, whichalways contains traces of calcium, copper, gold, and magnesium,and he points out the great importance of this in reference to theuse of the atomic weight of silver as a standard in other determina-tions of atomic weight.Jiolecular Weights.I n this section only one paper of importance has appearedduring the year, and in this is described the determination of themolecular weights of certain substances in bromine solution.26 Themethod adopted was t o aspirate air through the solution a t aknown temperature and t o estimate the bromine carried off by agiven volume of air by means of a solution of potassium iodide.The molecular weights were calculated from the formulaM=P’Sm/(P-P’).s, where ici and m are the molecular weights ofthe solvent and solute, S and s their weights in grams, P thevapour pressure of the pure solvent, and PI that of the solution.Antimony bromide, stannic bromide, and iodine monobromide werefound to be unimolecular, whilst a solution of sulphur consisted ofdiatomic molecules.25 A.de Gramont,, J . Chirn. phys., 1916, 14, 336 ; A . ii, 589.z6 R. Wright., T., 1916, 109, 113438 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Specific Heats.Reference was made in last year’s Report to the determinationof the specific heat of copper a t 15.2O and 21.6O abs.27 Thesemeasurements have now been repeated with a considerably im-proved apparatus,2* and the following values for the atomic heatof copper were obtained:Atomic heat.0.03960.11550-2340.870T. abs.14-51‘20.1925-3740-22If these values be compared with those calculated from the equa-tion C=kT3, it will be found t h a t the atomic heat decreases withfalling temperature more rapidly than it should do according toDebye’s law.The atomic heat of solid nitrogen has also beendetermined, and it is found to increase from 1.60 a t 15’27O abs. t o5.48 a t 61.68O abs. The variation of the atomic heat with tem-perature differs considerably from t h a t found with monatomicsolids, and therefore the conclusion is drawn that crystallisednitrogen consists of diatomic molecules.29The ratio of the two specific heats for various gases has beendetermined by Kundt’s method with tubes of different diametersand notes of different p i t ~ h . 3 ~ The results obtained would seemto show t h a t the molecular heats increase with the density of thegas. Thus, the following values were obtained for the molecularheats a t constant pressure : carbon monoxide, 6.900 ; nitrogen,6.905 ; oxyeen, 6.924 ; hydrogen chloride, 7.046 ; carbon chloride,8.904 ; water vapour, 9.214 ; sulphur dioxide, 10.059 ; ammonia,8.933 ; ethylene, 9.773 ; acetylene, 9.783.Allotropy.The conclusions drawn by Cohen and Helderman31 as to theexistence of two enantiotropic modifications of copper with a transi-tion temperature of 71-7O have been adversely criticised.32 Com-21 W.H. Keesom and H. K. Onnes, Proc. K . Akad. Wetensch. ,4msterdam.,1915, 17, 894; .4., 1915, ii, 83.28 Ibid., 1915, 18, 484 ; A., ii, 12.p9 Ibid., 1916, 18, 1247 ; .4.* ii, 371.30 G. Schweikert, Ann. Physik, 1915, [iv], 48, 593; 1916, [iv], 49, 433;31 E. Cohen and W. D. Helderman, Proc.I<. -4kad. Wetensch. -4msterdam,82 G. K. Burgess and T. N. Kellberg, J . Waphington Acad. 8 C L , 1916, 5,A , ii, 79, 216.1913, 16, 628; *4., 1914, ii, 205.657; A , , ii, 103.Ibid. 1914, 17, 60; A., 1914, ii, 654INORGANIC CHEMISTRY. 39parative measurements have been made with copper and platinumwire resistance thermometers wound on the same frame, and it isfound that platinum shows similar changes in conductivity t o thoseof copper when exposed alternately to temperatures of Oo andlooo. Both metals, however, give constant results after a fewalternations. A detailed study was made of the resistance ofcopper over the range of Oo to looo, but only negative resultswere obtained, and it is concluded that there is no evidence forthe existence of a metastable modification of copper.It must,however, be remembered that Cohen and Helderman based theirconclusions as to the allotropy of copper upon dilatometric measure-ments, and they only mentioned as collateral evidence the varia-tions in electrical resistance of copper noted by Matthiessen andBose after the copper had been kept some days a t a temperatureReference was made in the Report of 1914 to the new blackallotropic modification of phosphorus 34 prepared by heatingordinary yellow phosphorus a t 200° a t 12,000 kilogramsIcm.2Some further very interesting work on the allotropy of phosphorusmay be recorded. There is no doubt that black'phosphorus is adefinite allotropic modification and that red phosphorus is not adefinite substance, for it varies greatly in appearance and densityaccording t o the me'thod of preparation.35 The varieties close tothe upper limits of density are violet in appearance, and havebeen named by Smits violet phosphorus.This modification maybe prepared as follows. Ordinary white phosphorus with a traceof sodium as catalyst is subjected to a pressure of 4000 kilo-gramslcm.2 at the ordinary temperature, and then heated a t con-stant volume to 200O. The rise of pressure was only 500 kilograms.At 200° the pressure was raised to 12,500 kilograms for twentyminutes, and then t o 130,000 kilograms for forty-five minutes.No black phosphorus was formed. The apparatus was cooled, andthe pressure released a t the ordinary temperature. The white phos-phorus was entirely transformed into violet phosphorus, but a smallpiece of bright red phosphorus which had been placed under thewhite was unchanged.The density of the violet phosphorus pre-pared in this way was 2.348. A piece of violet phosphorus, togetherwith a small quantity of bright red phosphorus, was subjected t opressure and heat in the presence of iodine as catalyst. None ofthe black modification was formed, but the red phosphorus was34 P. W. Rridgman, J . Anzer. Chem. Sac., 1914, 36, 1344; A., 1914,35 Ibid., 1916, 38, 609; A4., ii, 246.of 1000.33Matthiessen and Bose, Ann. Phys. Chem., 1862, 115, 353.ii, 64740 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.converted into the violet modification. Evidently, therefore, theviolet modification is the stable form as compared with the redmodification at 200° between 8000 and 12,000 kilograms/cm.2 Itwas not found possible to convert red phosphorus into black phos-phorus, as a mixture of red and white phosphorus simply changedinto a mixture of red and black phosphorus when heated a t 200°and 12,500 kilograms/ cm.2An interesting observation was made on the rate of change ofwhite phosphorus to black phosphorus.After the pressure is raisedto 12,000 kilograms a t 200°, the black modification is never pro-duced instantaneously. There always is shown a period of pre-paration lasting from ten t o thirty minutes, during which periodthe pressure decreases slowly. The rate of decrease increasesgradually until a critical point is reached, when a sudden trans-formation of the entire mass to the black modification takes placeI n one experiment, the total drop of pressure during the pre-liminary stage was 400 kilograms in fifteen minutes, and duringthe transition the pressure dropped more than 8000 kilograms.Itis thus proved that the black modification is stable compared withwhite phosphorus a t 200° and pressures above 4000 kilograms.The violet modification is stable compared with the white modifica-tion a t 4000 kilograms and temperature somewhat below 200O.The violet modification is stable compared with red phosphorus a t200° between 8000 and 12,000 kilograms.Some attempts have been made to compare the vapour pressuresof black and violet phosphorus,36 but the results obtained were un-satisfactory, owing to the difficulty of removing the kerosene fromthe former, the kerosene having been used as pressure-liquid inits preparation.A t temperatures below 550°, the vapour pressureof black phosphorus did not reach a constant value. At some-what higher temperatures a constant value was reached, which isnearly equal t o that of violet phosphorus. At 570° the blackmodification has a higher vapour pressure than the violet modifi-cation. I n the presence of iodine as catalyst, the melting pointsof black and violet phosphorus were found t o be 587.5O and 589.5Orespectively.Smits’s theory of allotropy has been applied with some successto phosphorus.37 When violet phosphorus is rapidly and partlyvaporised in a vacuum a t a temperature a t which internalequilibrium is only slowly set up, a substance of abnormally low36 A.Smits, G. Meyer, and R. P. Beck, Proc. K . Akad. Welensch. Amster-37 A. Smits and S . C. Bokhorst, Zeitsch. physikal. Chem., 1916, 91, 249;dam, 1916, 18, 992; A . , ii, 185.A., ii, 317INORGANIC CH EMISTRT. d lvapour pressure is obtained. The more volatile component hasbeen expelled a t a temperature a t which it can only slowly bere-formed. The vapour pressure of this component increases con-tinuously a t constant temperature, and, after the addition of 0.1per cent. of iodine and heating a t 410°, the internal equiiibrium isagain set up and the vapour pressure becomes normal. Thisexplains the use of iodine in the melting-point determinationsmentioned above.It is concluded that red phosphorus is not anhomogeneous substance, and the name '' red-coloured phosphorus "is suggested.Colloids.An ingenious method has been tried with the view of determin-ing the relative quantities of free, capillary, and combined waterin inorganic gels.38 Since the total water content is known, itthen becomes possible to calculate the composition of the gel. Theprinciple of the method is a progressive freezing of the water,when the free and capillary water solidifies a t different tempera-tures. A weiphed quantity of the hydrogel is placed under lightpetroleum in a dilatometer, and the whole is slowly cooled. Thedilatometer readings a t first show a regular decrease, due t o thecontraction of the bulb and its contents.At a temperature usuallyseveral degrees below Oo, a sudden expansion occurs due to freez-ing. The water which freezes during this expansion consists offree water and that part of the capillary water the freezing pointof which has been reached. After the volume has become constant,a further lowering of the temperature causes more capillary waterto freeze, and the volume usually expands a little more. Afterthis the volume begins t o contract, and after the temperature hasbeen sufficiently lowered the contraction becomes a linear func-tion of the temperature. On raising the temperature again, thzexpansion is nearly linear for some distance, but a t a temperaturebelow Oo contraction sets in, due to the melting, and the volumefinally returns to its original value above Oo. It was not foundpossible sharply to distinguish between true capillary and freewater, for the reason that supercooling always occurs.Determina-tions, however, could be made of the amount of capillary waterfreezing below a certain temperature, and the temperature chosenwas -6O. The true capillary water is therefore rather greater inamount than this apparent capillary water.A number of preliminary experiments were tried with moistsand and moist lampblack in order to test the method and t oobserve the difference between free and capillary water. Sand88 H. W. Foote and B. Saxton, J . Amer. Chern. SOC., 1916, 38, 5 8 8 ; A.,ii, 230.c42 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.and water gave an example of a mixture containing only freewater. In this case, freezing occurred a t -4O, and the whole ofthe expansion took place at this temperature, the contraction aboveand below -4O being perfectly regular.The amount of waterpresent in the quantity of sand taken was 1.404 grams, and theamount estimated from the expansion was 1-430 grams.A sample of lampblack, which had previously been well ignited,was moistened with water and allowed to remain a few days.Although nearly dry to the touch, it then contained more t,han40 per cent. of water. The first expansion took place at -6O, butthe whole of the water was not frozen until the temperature hadremained a t - 2 8 O for two hours. The total quantity of water inthe sample taken was 1.834 grams and the total estimated was1.831 grams, and of this 1.207 grams was apparently capillarywater.Experiments were made with the gels of alumina, silica, andferric oxide, and also on the effect of ageing the gels by repeatedfreezing and remelting.The effect on the alumina gel is to causethe capillary water t o become free, whilst the silica gel reabsorbsthe capillary water after it has been frozen and remelted. Onecase of an alumina gel containing 51.90 per cent. of water wasfound t o have 37.1 per cent. in the combined state, the per-centage in Al(OH), being 34.6. Two gels of silica possessed com-bined water corresponding with the formu1z.e SiO2,l.31H,O andSiO2,l.35H,O, whilst a gel of ferric oxide contained combined wateragreeing with the formula Fe20,,4-25H,O.The interesting case of the doubly refractive sol of vanadiumpentoxide has been studied somewhat exten~ively.39~4~ I f a con-centrated solution of the sol is made t o flow through a tube oftriangular cross-section, which is used as a prism, it is found thatthe hydrogen red spectrum line is resolved into two oppositelypolarised lines.The more strongly refractive ray has its vibrationsparallel to the direction of flow, and therefore parallel t o themajor axes of the colloidal particles. When the solution is stirred,it exhibits yellow, shining streaks, and if a similar solution isexamined by transmitted light, it is seen to be quite clear, and darkstreaks are observed. The same phenomena can be seen if, insteadof stirring, the sol is placed in a magnetic or electric field.Whenthe sol is examined by the ultramicroscope,41 the presence is seen3B H. Diesselhorst and H. Freundlich, Phyeika'l. Zeitsch., 1915,16: 419 ;40 H. Freundlich, Zeitsch. E'lektrochem., 1916, 22, 27 ; A . , ii, 442.A., ii, 65.H. R. Kruyt,, Proc. K . Akad. Wetensch. Amsterdam, 1916, 18, 1625;A . , ii, 486INORGANIC CHEMISTRY. 43of very elongated, rod-like structures in quiet Brownian motion,together with small round disks in a state of rapid movement.When the sol is placed in an electric field and the force lines areparallel to the direction of the luminous beam, very little dis-persion takes place, and the elongated particles apparently dis-appear.On the other hand, when the field is perpendicular tothe luminous beam, the Tyndall effect is very pronounced. Allthe observations are in agreement with the view that the sol isvery similar to a liquid crystal.Many substances which act as protective colloids towards thehydrosols of silver, platinum, palladium, and similar metals areineffective in the case of colloidal mercury, for, owing to changeswhich take place on keeping, the mercury soon becomes insoluble.Albumins and their products of decompositions may, however, beused as protective colloids for the hydrosol of mercury,42 andstable, therapeutically active sols of mercury can then be prepared.If a solution of mercuric chloride be added to a solution of pyro-gallol, catechol, or certain aminophenols containing dextrin orglutin, a yellowish-white precipitate is obtained.On the additionof alkali, reduction takes place and colloidal mercury is formed.I n this way, solid hydrosols have been prepared containing up to80 per cent. of mercury.I f a mercuric salt is added to a solution of alkali containingdextrin or glutin, colloidal mercuric oxide is produced, and by thereaction between colloidal mercury and colloidal mercuric oxidestable preparations of colloidal mercurous oxide may be obtained.Rare Earths.Since the Report for 1912, no reference has been made t o anywork on the rare earths. It was thought preferable t o leave thisbranch of inorganic chemistry until the occasion offered of writinga short account of the work carried out during a period of severalyears. There is no doubt that the annual mention of papers onthis subject is unsatisfactory, because the lack of continuity therebyshown makes itself felt in this branch perhaps more than in anyother.During the interval of four years, a very considerable numberof investigations have been carried out on the rare earths, manyof which possess great interest.It is true that no startling dis-coveries have been made, and the majority of the papers thatcome under review deal with technique, that is t o say, t,he methodsof separating these elements the one from the other. Our know-ledge of the salts of the rare earths, their properties and solubili-4a C. Amberger, Kolloid Zeitsch., 1916, 18, 97 ; A., ii, 380.c * 44 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.ties, has greatly been enhanced.It is only right to say that thisadvance is due in the main to the work of Professor C. James andhis co-workers. So much is this the case that it is convenient forthe sake of greater continuity to discuss his papers separately.I n the following review, the first three investigations describeddeal with the general properties of compounds of the rare earths,whilst the later papers describe work that has been carried outon the solubilities of their many salts and the methods that havebeen experimented with for the purpose of effecting the separationof the rare earths from one another.When the sulphates of the rare earths are heated, they lose SO3and are converted into basic sulphates, which are insoluble inwater. The dissociation pressures of some of the sulphates havebeen measured a t temperatures between 800° and 1200O with theview of obtaining a measure of their basic affinity.@ I n thefollowing table are given the values of the dissociation pressure (P)of the sulphates in mm.of mercury a t 900°, and also the heats ofdissociation (Q) calculated according to Nernst's theory :Sc. Sa. Gd. Nd. Pr. Er. Yb. Lu. Yt. La.P 11.0 8.0 7-0 6.0 5.5 5.0 4.0 3.5 3.0 2.0Q 54-5 56-6 56.9 57-2 57.4 57.6 58.2 58.5 58.9 59.8It is evident from the above figures that the rare earths belong tothe strongest bases. It is unfortunate that the dissociation pressurecurves run almost parallel and very close to one another, so thatit would, in general, be very difficult to separate any two rareearth sulphates by heating them t o a constant, intermediatetemperature, at which one gives the basic sulphate and the otherremains undecomposed and soluble in water.The ethyl sulphates 44 have been prepared of yttrium, lanthanum,cerium, praseodymium, neodymium, samarium, europium, gado-linium, dysprosium, thulium, erbium, and neoytterbium.Thesesalts have the general formula M2(EtS0,),,18H,0, and they arestrictly isomorphous, crystallising in the hexagonal system.Although the corresponding salts of scandium and indium wereobtained in very small crystals, yet i t was possible to concludethat they have not the same form, being monoclinic. Glucinuniethyl sulphate stands in quite a different' class from the other salts,for it is a basic salt with the formula G10,G1(EtS0,),,4H20.The acetonylacetates of scandium, indium, and iron are iso-morphous, and they form large, flat crystals quite different fromthe fine needles in which the corresponding salts of the rare earthsl3 L.Wohler and M. Griinzweig, Ber., 1913, 46, 1726; A . , 1913, ii, 597.44 F. RI. Jaeger, Proc. K . d k a d . Wetensch. Amsterdam, 1914, 16, 1095 ;A., 1914, i, 797INORGANIC CHEMISTRY. 45separate out. Glucinum acetonylacetate is again quite differentfrom all the other salts.I n the literature dealing with neodymium sesquioxide, variouscolours have been attributed to it, such, for example, as blue,ash-grey, green, pale violet, lavender. An investigation has shownthat these contradictory statements are due to the fact that thereexist hydrated oxides with different colours.45 When neodymiumhydroxide is heated a t 320°, the hydrated oxide, 2Nd20,,3H,O, isformed. This hydrate is stable up to about 500°, and has a palebrown colour tinged with rose.It heated a t about 550°, it isconverted into the hydrated oxide, Nd,O,,H,O, the colour ofwhich is less pronounced. Neodymium sesquioxide is only formedquantitatively at 10GOo, and is blue. The various coloursdescribed by previous investigators are evidently due to their hav-ing obtained indefinite mixtures of the sesquioxide and the twohydrated oxides.An interesting account has been published of the fractionationof the rare earths of the ytterbium group with the view ofestablishing once and for all the homogeneity of the element neo-ytterbi~m-~e The method adopted was the fractional crystallisa-tion of the nitrates, the operation being followed by measuringthe coefficients of magnetisation of each fraction.After 4000fractionations, 8 successive fractions were obtained, eachhaving the same coefficient. This established quite definitely theelementary character of neoytterbium. As previously reported,the atomic weight was found to be 173.54. The arc spectra weremeasured of the two extreme fractions of the series of 8, andwere found to be identical, except for a few lines in the one duet o thulium and a few in the other due to luteciurn. None of therays attributed to aldebaranium 47 was seen ; indeed, considerabledoubt has been thrown on the existence of this as a definiteelement.The glycollates of certain of the rare earths have been preparedand their properties described.48 The glycollates of the earths ofthe cerium group are anhydrous and crystallise in crusts, whilstthose of the earths of the yttrium group crystallise in needles, withtwo molecules of water of crystallisation.The yttrium salt isthe least soluble, then follow the lanthanum, cerium, and praseo-dymium salts, which are almost equal, and then, in order, the45 C. Gamier, Arch. Sci. phys. nut., 1915, [vi!, 40, 93, 199; A., 1915, ii, 775.46 J. Blumenfeld and G. Urbain, Compt. rend., 1914, 159, 323, 401; A.,47 C. A. von. Welshach, Monatsh., 1908, 29, 181 ; A., 1908, ii, 591.1914, ii, 731, 694.G.Jamtsch and A. Griinkraut, Zeit.sch. arcorg. Chem., 1913, 79, 305;A., 1913, i, 24746 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.neodymium, samarium, and gadolinium salts. Measurements ofthe electrical conductivity of the solutions show, however, t h a tcomplex salts are present. When lanthanum hydroxide is warmedwith a solution of glycollic acid, it dissolves to form a clear solu-tion, b u t a t a definite temperature, which depends only on theconcentration, a complex salt separates out as a precipitate. Thepraseodymium, neodymium, and samarium salts behave in asimilar way.crystallises without forming an unstable solution, b u t the yttriumsalt behaves like the above.A trial was made of using these gycollates as a means offractionating the rare earths from xenotime, after they had beenfreed from cerium.Successive fractions showed a progressiveincrease in atomic weight, and the spectroscope revealed aconcentration of neodymium and praseodymium in the lastfractions.When neutral solutions of the nitrates of the rare earths areelectrolysed with a mercury cathode and platinum anode, thehydroxides are precipitated, the velocity of precipitation depend-ing on the basicity of the rare earths.49 Some experiments weremade in order to test whether this method would be effective i nseparating the rare earths from one another. When a neutralsolution of the nitrates of neodymium, praseodymium, lanthanum,and samarium is electrolysed, the lanthanum collects in the lastfractions, and can thus be separated from the other earths of thedidymium group.Experiments have also been made on the fractional electrolysisof the rare earths from xenotime, and it was found that erbiumcollects in the early fractions and yttrium i n the later fractions.The general results of the investigations indicate that the separa-tion of some of the rare earths can be effected by this methodmore rapidly- and conveniently than by the usual methods offractional crystallisation and precipitation.It was thoughtpossible t h a t the precipitation of the hydroxides might be due tothe production of ammonia by reduction of the nitric acid. I norder to test this point, similar experiments were carried out withthe chlorides of the rare earths, and analogous results wereobtained, b u t the rate of precipitation was found to be muchslower than that observed in the case of the nitrates. The nitrateexperiments were repeated, using a diaphragm, and in one case,with a solution rich in erbium, holmium, thulium, and yttrium,On the other hand, gadolinium gycollate,Gd( CzH,03)3,2H,O,4s L.M. Dennis and R. J. Lemon, J. *4mer, Chena. IcIoc., 1015, 37, 131,1963 ; A . , ii, 99, 77IXORGANIC CHEMISTRY. 47the first three elements were not appreciably separated from oneanother, b u t they were rapidly separated from the yttrium.The method of fractional crystallisation of their picrates as ameans of separating the rare earths from one another was firsttried in 1912.50 This method has now been studied anew.51 Amixture of the picrates of the rare earths of the didymium group,containing small quantities of those of the yttrium and erbiumgroups, was submitted to 42 series of fractional crystallisations.I t was found that praseodymium and neodymium tend t o con-centrate in the least soluble fractions, t h a t yttrium concentratesin the most soluble fractions, whilst the metals of the erbiumgroup are found in the intermediate fractions.In general, thismethod does not effect a separation of neodymium from praseo-dymium, but it is useful for removing small quantities of theearths of the erbium and yttrium groups from those of thedidymium group, and also for the separation of yttrium fromerbium and holmium.Finally, in concluding this section of this review, a short refer-ence may be made to a paper describing the separation from oneanother of the rare earths from gadolinite and xenotime.52Various methods of fractionation were employed.The fractionsobtained by the bromate method were submitted t o further frac-tionation by the chromate method. In this way, small quantitiesof yttria can rapidly be prepared from mixtures containing smallamounts of erbium and holmium, but the method cannot be appliedto the fractionation of mixtures which contain didymium andgadolinium unless these earths are first removed by means ofpotassium sulpha t e.A rapid survey may now be given of the work of James and hisco-workers, and in the first place brief mention may be made ofthe fractionation of the rare earths derived from the Carolinamonazite sands.The bromate method of fractionation was used.53The sands contain lanthanum, cerium, praseodymium, neodymium,together with considerable quantities of samarium, gadolinium,yttrium, small amounts of dysprosium, holmium,, erbium, andminute amounts of europium, terbium, thulium, ytterbium, etc.It is not possible t o give a detailed account of the work, but thisbrief mention will show t h e complexity of the problem.An important point is raised with regard t o the estimation ofthe rare earths, for it is found that, when they are precipitatedL. M. Dennis and C. W. Bennett, J . Amer. Chem. "3oc.,1912, 34, 7 ; &4.,1912, ii, 257.61 L.M. Dennis and F. H. Rhodes, ibid., 1915, 37, 807 ; A . , 1915, ii, 347.52 J. R. Egan and C. W. Balke, ibid., 1913, 35, 365 ; A . , 1913, ii, 508.53 C. James, ibid., 1913, 35, 235 ; A . , 1913, ii, 32348 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.as hydroxides and ignited to the oxides, the results obtained arehigher than when they are precipitated as oxalates and ignited.The explanation of this is t o be found in the fact that the hydr-oxides tend always to carry down some of the alkali when theyare precipitated, and especially is this the case when sodiumhydroxide is used as p r e ~ i p i t a n t . ~ ~I n previous papers it had been shown that yttrium can beseparated quantitatively from the alkali metals by precipitationwith ammonium sebacate.It i s now found that lanthanum andcerium can also be separated in the same way.55Sodium cacodylate as precipitant has also been experimentedwith as a method of separation of the rare earths. On fractionallyprecipitating a solution containing chiefly the chlorides of yttrium,dysprosium, and holmium with sodium cacodylate, the yttriumtends t o accumulate in the early fractions, and the holmium anddysprosium in the later fractions. On boiling a mixture of thehydroxides of neodymium, samarium, and gadolinium withcacodylic acid, and fractionally crystallising the cacodylates fromhot water, neodymium collects in the more soluble fractions, whilstnearly all the terbium and dysprosium remain in the least solubleportions. I n this paper are described the methods of preparationand the properties of the sebacates and cacodylates of a numberof the rare earths.I n the separation of the earths of the cerium and yttriumgroups, the use of sodium sulphate has led to varying results.This uncertainty has been investigated, and it' is found that theefficiency of sodium sulphate depends on its concentration.66 Ifthe solution is too concentrated, much of the yttrium earths isalso precipitated.As regards the quantitative separation of neodymium fromglucinum, titanium, barium, and uranium, it is found that thiscan be effected by the precipitation of the neodymium as oxalatefrom the boiling solution.57 Oxalic acid is used, and, after precipi-tation, the solution is digested until the precipitate becomesgranular, when it mag easily be filtered.The preparation of the dimethyl phosphates of a number of therare earths are described in another paper.58 The solubilities ofthese saIts have been determined, and i t was found that they aremore soluble in cold water than in hot.A trial was made of the54 T. 0. Smith and C. James, J. Amer. Chem. SOC., 1914, 36, 909; A . ,56 C. F. Whitternore and C. James, ihid., 1913, 35, 127 ; A., 1913, i,?48.66 C. James and H. 0. Holden, ibid., 559; A . , 1913, ii, 508.67 T. 0. Smith and C. James, ibid., 563 ; A., 1913, ii, 531.ss J. C. Morgan and C. James, ibid., 1914, 36, 10 ; A., 1914, i, 135.1914, ii, 492INORGANIC CHEMISTRY. 49possibilities of this method for the Separation of the rare earths.Fractionation experiments were made by preparing a solution ofthe rare earths in dimethylphosphoric acid.The temperature ofthe solution is gradually raised, and the precipitates formed arecolIected at definite temperatures. By the evaporation of themother liquor additional fractions are obtained. It was foundthat the rate of separation of the rare earths by this method ismuch greater than by other methods. Lanthanum, cerium,praseodymium, and neodymium are left in the mother liquor.Samarium, europium, and gadolinium are much less soluble thanthese, but they are more soluble than terbium, dysprosium, andholmium. As a result of this fractionation, erbium, thulium,yttrium, and ytterbium collect in the least soluble fractions. Itmust be pointed out, however, that the rare earth dimethyl phos-phates are inclined to decompose, with the formation of gelatinousprecipitates, which render the filtration somewhat difficult.A great number of investigations have been made in order t ofind the best method for the separation of yttrium from theyttrium earths.The following methods were tried : the fractionalprecipitation of the phosphates, monomethyl phosphates, dimethylphosphates, cacodylates, arsenates, phosphites, chromates, bromates,and iodates, and also the fractional precipitation by means ofhypophosphorus acid.59 It was found that the chromate andphosphate methods gave the greatest efficiency. It would seem,however, that the best method is to use sodium nitrite. The rareearth oxides are dissolved in nitric acid, and the solution is dilutedand boiled.A quantity of sodium nitrite is then added, whichis sufficient to precipitate the required fraction o l the rare earthmaterial. The yttrium concentrates in the later fractions. Thismethod gives a larger yield and is less expensive than either thephosphate o r the chromate method. It is not very effective, how-ever, for the separation of yttrium from terbium.I n addition to the above, various other methods f o r separationby fractional precipitation were tried.60 It appears that potassiumcobalticyanide is one of the most promising reagents. The rareearth cobalticyanides separate out in crystals, which have thegeneral formula h~2(CoC,N6)2,9H,0.61 Measurements were madeof the solubilities of these compounds in 10 per cent.hydrochloricacid (D15 1.050), and the following results were obtained, expressedin parts of the salt contained in 1000 parts of the saturated solu-H. C. Holden and C. James, J . Amer. Chem. Soc., 1914, 36, 634, 1418 ;A., 1914, ii, 370, 657.60 J. P. Bonardi and C. James, ibid., 1915, 37, 2612 ; A . , ii, 102.C. James and P. S. Willancl, ibid., 1916, 38, 1497 ; A . , i, 63850 BNNUAL REPORTS ON THE PROGRESS O F CHENISTRY.tioa : lanthanum, 10.41 ; cerium, 10.75 ; neodymium, 4-19 ; gado-linium, 1.86 ; yttrium, 2.78 ; ytterbium, 0.38. Fractional crystal-lisation experiments showed that- a rapid separation may beobtained by the cobalticyanide method, and its use is particularlyrecommended for the separation of yttrium from erbium. On thelarge scale, fractional precipitation by means of sodium nitrite isrecommended, on account of the cheapness and the ease ofin anipula tion.Owing t o the fact t h a t some doubt had been thrown on thehomogeneity of terbium as an element, a long series of fraction-ations was undertaken in order, if possible, to decide the clues-tion.62 The material employed consisted of gadolinium oxidecontaining terbium oxide, together with dysprosium and neo-dymium oxides and traces of the oxides of yttrium and erbium.From this mixture the bromates were prepared, and these weresubmitted to a long and careful fractionation.The results obtainedleave no doubt t h a t terbium is a single, homogeneous element.The bromate process was found to effect a comparatively rapidseparation of terbium from gadolinium, and neodymium, if present,comes between these two.Finally, reference may be made t o a paper in which a verycomplete account is given of the separation of the rare earths, inwhich many of the above methods are employed.63 The value ofthis paper lies in the fact t h a t examples are given of the applica-tion of the methods to actual practice.The source of the rareearths was Brazilian monazite sand, and the material used was thesolution obtained after potassium sulphate had been added to themixed rare earth sulphates in quantity insufficient to precipitatecompletely the whole of the cerium metals. The solution containedconsiderable quantities of lanthanum; cerium, praseodymium,neodymium, in addition to samarium, gadolinium, and the earthsof the yttrium group.Reference is made t o the difficulty ofisolating holmium, and it is said t h a t this is one of the mostdifficult problems of inorganic chemistry.I n the above account of James’s work, only the more importantpapers have been referred to, and it is hoped t h a t sulficient hasbeen said t o make the story comprehensible and to create a highopinion of the value of this work.ii, 811.82 C. James and D. W. Bissel, J . Amer. Chem. Soc., 1914, 36, 2060 ; A.,Ba C. James and A. J. Grant, ibid., 1916, 38, 41 ; A4., ii, 251INORGANIC CHEMISTRY. 51Group I .It was shown a few years ago that ammonium chloride andammonium bromide are enantiotropic, with transition tempera-tures at 159O and 1090 respectively.64 Both these substances havebeen reinvestigated.I n general, the cooling-curve method givestoo low a value for the transition temperature, owing to the slow-ness a t which equilibrium is established. I n the case of ammoniumchloride, it has been found65 that if glycerol or mannitol is usedas catalyst, the method gives excellent results, since the transitiontemperature given by the cooling and heating curves are in fairlygood agreement. The solubilities of the salt in water betweenthe temperatures 160° and 205O have been determined, and fromthese the transition temperature has accurately been calculated.If x is the number of molecules of the salt dissolved in 1 moleculeof the saturated solution, .then, by plotting log x against 1 IT, twostraight lines are obtained, intersecting a t 184'5O, which is there-fore the true transition temperature for ammonium chloride.Thesolubility of the modification stable at lower temperatures is givenby the relation - log x = 464.5 / T - 0-5400, whilst that of the modifi-cation stable a t higher temperatures is given by-log ~=327.8/T-0*2412.I n the case of ammonium bromide, the thermal method showsthat the transition temperature lies between 130° and 143O.66 Theexact transition temperature was obtained by the determinationof the solubilities of the salt in water between Oo and 170O. Thesolubility curve shows a well-defined break a t 137.3O, which istherefore the true transition temperature of ammonium bromide.The heat change involved in the transition of ammonium chloridehas been determined by the calorimetric method,67 in whichweighed quantities of the salt, heateci t o known temperatures aboveand below the transition temperature of 184'5O, were introducedinto a caIorimeter and dissolved in the water contained therein.The heat change iiivolved in the transformation of the a- into the&form is -1030 calories. The theory was put forward byWegscheider that the dissociation of ammonium chloride vapouris due to the existence of two polymorphic forms, one of which64 R.C. Wallace, Centr. Min., 1910, 33 : A . , 1910, ii, 208.66 F. E. C. Scheffer, Proc. K . Akad. Wetensch. Amaterdum, 1915, 18, 446 ;66 A.Smith and H. E. Eastlack, J. Amer. Chem. Soc., 1916, 38, 1261 ;F. E. C. Scheffer, Proc. K . Akad. Wetensch. Amsterdam, 1916, 18, 1498 ;A., ii, 31.A., ii, 483.A., ii, 43158 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.passes into the other under the influence of water, This theoryhas now been disproved.The case of ammonium iodide presents some interest, becausethe crystalline habit of this salt differs from that of the bromideand chloride. It would appear probable that ammonium iodidetherefore also exists in two enantiotropic modifications as well asthe other two salts. Solubility determinations, however, between- 1 9 O and 136O have failed t o reveal the existence of a transitiontemperature .6*A new method has been described 69 of separating rubidium andcaesium, based on the solubilities of their respective alums incold water.A t 15--17O, 100 grams of water dissolve 2.3 gramsof rubidium alum and 0.62 gram of czsium alum. I n order toobtain these two elements from lepidolite, the mineral is decom-posed by heating it with calcium fluoride and concentratedsulphuric acid, and the resulting solution is freed from calciumsulphate and evaporated until the mixed alums crystallise. Thecrystals are separated and recrystallised, the mother liquor beingadded to the mother liquor from the first crop. The mixed liquorsare then evaporated t o crystallisation, and so on. After thirteenoperations, the first crystals which separate are pure caesium alum,but twenty-eight operations are necessary in order to obtain purerubidium alum.If the mixed alums are dissolved in water, thealuminium hydroxide precipitated by the addition of ammonia,the filtrate evaporated and treated with ammonium ferric alum,the crystals which first separate are caesium alum, and are freefrom rubidium, I n order to separate cmiurn from pollucite, themineral is decomposed with hydrochloric acid, the silica is filteredoff, and the acid filtrate is warmed after the addition of ammoniumaluminium alum. On cooling, crystals of caesium alum separateout, and after two recrystallisations a pure salt is obtained. Theremainder of the caesium may be obtained by a few recrystallisa-tions of the mother liquor.Some interesting work may be recorded on the nitrogeneouscompounds of gold.70 These may not only be obtained from thesalts and oxides of gold and aqueous ammonia and ammoniumcarbonate, but also from auric hydroxide and the ammonium saltsof strong acids.I n general, the compounds are very explosive,and are decomposed by washing with water. It was therefore68 A. Smith and H. E. Eastlack, J . Amer. Chern. SOC., 1916, 38, 1500;A., ii, 529.a g P. E. Browning and S. R. Spencer, Amer. J . Sci., 1916, [iv], 42, 279 ;A . , ii, 563.7 0 E. Weitz, Annalrzn, 1915, 410, 117; A., ii, 39INORGANIC CHEMISTRY 53found necessary to analyse them in tthe wet condition. The follow-ing details may be given. If aqueous solutions of chloroauric acidare treated with five or more molecules of ammonia, precipitat,esare obtained.As the amount of ammonia is increased, thequantity of chlorine in the precipitate decreases, but the ratioAu : N remains constant a t 1 : 1.5. These precipitates are mixturesof two compounds,Au,O53NH3, sesquiammineauric oxide ;NH(AuCl,NH,),, diaminoiminodiauric chloride.The compound, 2Au(OH),,3NH3, has also been obtained. Whendried in air, this compound is relatively harmless, but after dry-ing a t 105-110° or in a vacuum over phosphoric oxide, it loseswater and becomes very explosive. Other compounds have beenprepared with the same ratio Au: N, namely, 1 : 1.5.If N / 5-chloroauric acid, containing ammonium chloride, is addedto a cold saturated solution of ammonium chloride saturated withammonia, a dense yellow precipitate is formed.This compoundhas the formula Au(NH,),Cl,H,O, and has been named diamino-auric chloride, and it is not explosive. On washing with water,however, it is converted into an explosive substance, which possiblyhas the formula 3Au0,2NH3,yH,0. Diaminoiminodiauric chlorideand diaminoauric chloride each give, on continued treatment withaqueous ammonia, hydrated sesquiammineauric oxide. This substance, on heating a t 115-120°, gives a very explosive blacksubstance, diamminetriaurous oxide, 3Au,0,2NR3, and with hotwater it gives a still more explosive monoammineauric oxide,Au,0,,2NH3. I f N / 5-chloroauric acid, saturated with ammoniumnitrate, is added t o a cold saturated solution of ammonium nitrate,and the mixture saturated with ammonia a t the ordinary tempera-ture, a precipitate is formed, which can be recrystallised from warmwater.It forms colourless needles, and is tetra-ammineauric nitrate,[AU(NH~)J(NO~)~. This compouud forms double salts withpotassium, sodium, and ammonium nitrates. By double decomposi-tion, several other salts of the radicle Au(NH3), have been pre-pared, but the hydroxide and the haloid salts have not beenobtained.The ammoniates of certain inorganic salts of silver have beendescribed.71 Three different methods were made use of in theirpreparation, namely, the absorption of ammonia by the dry saltsat different temperatures, the precipitation by ether of ammoniacal-alcoholic solutions of the silver salts, and the evaporation of aqueousammoniacal solutions of the salts.The following compounds have'l Q. Bruni and G. Levi, Cfazzetta. 1916, 46, ii, 17, 235; A., ii, 482, 61754 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.been prepared : Ag,0,2NI-13, AgF,2N11,,21-120, AgC103,3N13,,AgC1,,2NH3, and 3NH3,AgBr0,,3NH, ; AgMn0,,3NH3,Ag,S03,4NH3, Ag,Se0,,4NH3, and NH,S03Ag,2NH,.Pure cuprous sulphide prepared in the vacuum furnace has amelting point of 1130+1°, and it does not dissociate up t o atemperature of l2OOO.72 It has DY 5.785, a value which is almostidentical with that of the purest mineral sulphide. The sulphidesformed by fusing together copper and sulphur are o€ variablecomposition, and always contain more sulphur than correspondswith the ratio 2Cu:S. They are, in reality, solid solutions ofcupric and cuprous sulphides.Cuprous sulphide in an atmosphereof hydrogen sulphide melts a t 1096O, and at 1057O in an atmo-sphere of sulphur vapour. This is due to the formation and solu-tion of cupric sulphide. When cuprous sulphide is heated in anatmosphere of hydrogen sulphide at various temperatures, thesulphur content increases, there being a definite sulphur contenta t each definite temperature. The sulphur content increases withdecrease of temperature until a t 358O cupric sulphide is formed.Cuprous and cupric sulphides when heated together give similarsolid solutions, and these have been found in nature. Cuproussulphide is dimorphous, with an inversion temperature a t 9 1 O .The crystals formed by the action of ammonium sulphide onmetallic copper are not cuprous sulphide, but a double sulphideof the formula Cu,(NH,)S, or 7Cu,S,(NH,),S.Group 11.A method has been described for the extraction of glucinumfrom gadolinite.73 The mineral, after it has been crushed andpowdered, is decomposed with concentrated sulphuric acid.Thesulphates are dissolved in water and separated from the silica bydecantation. The rare earths are then precipitated by oxalic acid,and the iron and glucinum in the filtrate, after the oxalic acidhas been oxidised, are separated by fractional precipitation of theirhydroxides by means of sodium hydroxide. The two metals arc3first precipitated together, and then the hydroxides are stirredwith sufficient acid t o dissolve about two-thirds of the whole.Allthe glucinum is thus dissolved, together with some iron, and thiscan be almost entirely precipitated by means of more sodium hydr-72 E. Posnjalr, E. T. Alleii, and 13. E. Merwin, Economic GeoZogy, 3915,79 C. James and G. A. Perley, J . Amer. Chem. Soc., 1916, 38, 875; A.,10, 491 ; A., ii, 103.ii, 326INOIiGANIC CHEMISTRY. ti 5oxide. Tho last traces of iron are removed by means of hydrogensulphide, and the glucicum is precipitated as the basic carbonate.The absorbent power of metallic calcium for gases is well known,and was first used by Soddy as a means of producing high vacua.7iIt appears that there are two forms of metallic calcium, an activeand an inactive form.75 The active modification begins to absorbnitrogen a t 300°, and the velocity of the reaction increases withthe temperature until it reaches the maximum a t 440°, above whichtemperature the velocity decreases until it vanishes a t 800'.Thevelocity depends on the presence of a layer of the nitride, and onlyreaches its maximum value after this layer has been formed.The inactive form only commences to combine with nitrogen a t800O. These two forms of metallic calcium do not appear to beallotropic modifications, but merely the metal in two different; statesof subdivision. When melted calcium is slowly cooled, the activeform is produced, and this gives a brown nitride. The inactivcform is produced by suddenly cooling calcium from 840°, and itgives a black nitride. The active form absorbs hydrogen between150° and 300°, and above 600° calcium nitride absorbs hydrogen,carbon monoxide, carbon dioxide, and methane.When a mixture of potassium and magnesium chlorides iselectrolysed in a graphite crucible, small quantities of a blackcompound are found in the mass after cooling.76 This substanceevolves hydrogen when treated with water, and does not precipi-tate metallic nickel from an anhydrous solution of nickel chloridein alcohol, a reaction which is shown by metallic magnesium.Itis not formed during the electrolysis in complete absence of oxygen,and it is believed to be magnesium suboxide.I n addition to the two well-known forms of calcium carbonate,calcite and aragonite, a third form has been described, to whichthe symbol pCaCO, has been given.77 It, is best obtained by pre-cipitation a t 60°, but it is always contaminated with calcite oraragonite; it may, however, be separated by flotation in a liquidD 2.6.The substance has D 2.54, those of calcite and aragonitebeing D 2.71 and D 2-88 respectively. It forms microscopic platesbelonging to the hexagonal system.The existence of calcium hydrogen carbonate as a definite com-pound in aqueous solution has been proved.78 The maximum74 F. Soddy, Proc. Roy. JCOC., 1907, [A], 78, 429 ; A., 1907, ii, 251.75 A. SSieverts, Zeit.scJz. Elektrochem., 1916, 22, 15 ; A., ii, 432.76 F. C. Frary and A. C. Berman, Trans. Amer. Electrochem.. SOC., 1915,77 J. Johnstone, I€. E. Mermin, and E. D. Williamson, Amer. J . Sci.,27, 209 ; A., ii, 33.1916, [iv?, 41, 473 ; A . , ii, 433.A.Cavazzi, Gazzetta, 1916, 46, ii, 122; A . , ii, 53056 ANXUAL REPORTS ON THE PROGRESS OF CHEMISTRY.quantity of calcium carbonate, which dissolves on shaking for tenhours at Oo in 1 litre of water saturated with carbon dioxide andmaintained so a t atmospheric pressure, is 1.56 grams, which corre-sponds with 2.5272 grams of calcium hydrogen carbonate. Underthe same conditions a t 15O there is dissolved 1.1752 grams ofcalcium carbonate, which corresponds with 1'9028 grams of calciumhydrogen carbonate. When carbon dioxide is passed very rapidlyinto lime-water saturated at 15O, the solution finally becomes clearand forms an unstable solution, supersaturated with the gas andcontaining 2.29 grams of calcium carbonate or 3-71 grams ofcalcium hydrogen carbonate in 1 litre.Although perhaps outside the purview of this Report, yet itcannot be denied that the phosphorescence of inorganic compoundsis a phenomenon of peculiar interest to the inorganic chemist,especially since it has been proved that no pure substance phos-phoresces.Phosphorescence, when exhibited, is always due to thepresence of an impurity, known as the phosphorogen, and thefollowing directions have been given for the preparation of a phos-phorescent calcium sulphide.79 A mixture of 100 parts of calciumcarbonate and 30 parts of powdered sulphur is heated at a dullred heat for one hour. It is then cooled and mixed with alcohol,and sufficient of an alcoholic solution of basic bismuth nitrate isadded so as to give 1 part of bismuth to 10,000 parts of calciumsulphide.The mixture is dried in air and then heated at a dullcherry-red heat for two hours, after which it is slowly cooled. Thebismuth, as phosphorogen, may be replaced by molybdenum,uranium, or, best, by tungsten.Group 111.The action of nitric acid on metallic aluminium under varyingconditions has been investigated.*" The principal conclusionsdrawn from the work are as follows: The most important factoris the temperature, for over a considerable range an increase ofloo is sufficient t o raise the rate of dissolution by 100 per cent.Next to the temperature, the concentration plays the most pro-minent part. The most active solvents are mixtures containingbetween 20 and 40 per cent.by volume of nitric acid (D 1-42). Onthe other hand, acids containing 94.7 per cent. nitric acid arealmost without action. The presence of chlorine up t o 0.05 percent. in the nitric acid dces not affect the rate of attack, but tracesof sulphuric acid materially increase it. I n one case i t was found7 9 P. Rreteau, Compt. rend., 1915, 161, 732; L4., ii, 100.8o R. Seligmzn and P. Williams, J. SOC. Chem. Ind., 1916, 35, 6 6 5 ; A . ,ii, 435INORGANIC CHEMISTRY. 57t h a t 0.04 per cent. ?ulphuric acid was sufficient to increase theamount of aluminium dissolved from 36 to 62 mg. per 100 sq.cm.per twentv-four hours. The presence of t h e lower oxides of nitro-gen accelerates the action, and again the metal is more readilyattacked when amorphous than when crystallised.Mixtures ofnitric and sulphuric acids attack aluminium very much morereadily than pure nitric acid.Many hydrates of aluminium nitrate are described in the litera-ture beyond the one containing 18H20. For example, a hydrstehas been described with 15H20, which is stated to be quite stablein tlie air.81 I n the work mentioned above, the solutions obtainedoften became sa,turated with aluminium nitrate, and gave copiouscrops of crystals. Three definite hydrates were recognised 82 with18H20, 15 or lGH,O, and 12H,O, which differ appreciably in habit,stability, solubility, and vapour pressure. When aluminium iskept in contact a t 20° with a limited amount of nitric acid (D 1.42)very slow dissolution occurs, and the solution rapidly becomessaturated with aluminium nitrate.Large, clear, colourless crystalswere deposited, but, during the halidling of the supernatant liquor,it became clouded by a mass of small needles, which, on touchingthe original cryst&, changed them also into needles. The stableform of the crystals was proved t o be tlie known hydrate containing1 8H,O. Saturated solutions of aluminium nitrate, made by lieat-ing the metal with nitric acid (D 1-42) and filtering through glass-wool, readily deposited the unstable crystals. These were provedt o be a new hydrate containing 12H,O. The third hydrate wasprepared as follows: The mother liquor froin the hydrate contain-ing 12N,O gave, on keeping, very thin plates, some of which werevery nearly 1 sq.cm. in est'ent. These crystals were found tocontain either 15H,O or 16H,O, but it was difficult to decideexactly the number of molecliles owing t o the impossibility ofcompletely drying the crystals.The alumina produced by the hydrolysis of aluminates is differ-ent in its nroperties from tlist precipitated cn the addition ofammonia to aluminium salts in solutioii.s3 It is of a sandy natureand is non-hygroscopic. It has the composition AI20,,3H,O, and,when heated a t 225O, it is converted into A1,03,2H20, and a t 2 3 5 Oit becomes Rl,O,,H,O. The hydroxide precipitated from alumin-ium Salts, when dried a t 80°, is approximately A1,0,,2H20. Onheating further it loses water, and a t no temperature does itappear to give a definite hydrate.M.2. Jovitschitsch. Monatslz.., 1912, 33, 1 6 ; A., 1912, ii, 26182 R. Seligman and P. TVilliams, T., 1916, 109, 612.sn E. Martin, il4on. Sci., 191.5, [vj, 5, 2 2 5 ; A., ii, 13958 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.When mixtures of aluniinium oxide and barium carbonate in anyproportions are heated a t temperatures not exceeding 1500°, thealuminate, AI,O,,BaO, is always formed. A t the temperature ofthe electric arc, tribarium aluminate, A1203,3Ba0, is obtained froma mixture of 1 mol. of alumina and 3 mols. of barium carbonate.This compound is soluble in water. Certain other compounds aredescribed, with the following conipositions : 10A1,0,,1 1Ba0,55H20 ;A120,,2Ca0 ; A1,03,2BaS ; 2A1203,3BaS.In last year’s Report it was stated that no perborate is formedwhen solutions of sodium and potassium borates, with varyingproportions of alkali hydroxide, are electrolysed.84 It appears,however, that sodium perborate may readily be prepared by theelectrolysis of a solution of borax containing sodium carbonate.85A solution of 40 grams of borax and 120 grams of anhydroussodium carbonatel is electrolysed between a platinum gauze anodeand a tin tubme cathode, which is bent round tlie anode.By thecirculation of water through the cathode the temperature, is kepta t 18O. Using a current of 20 amperes a t 6 volts, with an anode8 x 6 cm., as much as 20 grams of sodium perborate were obtainedin one hour.Group IV.A study has been made of the graphitic acids produced fromvarious graphites.86 The use of permanganic acid is unsatisfactory,as it gives graphitic acids of variable composition.It is prefer-able to use a mixture of concentrated sulphuric and nitric acidswith potassium chlorate. There are two graphitic acids in leaflets,which differ in their colour and composition. Acheson, Ceylon,and Russian blasbf urnace graphites give yellow graphitic acid,whilst those from Siberian, Italian, Corean, and cast-iron graphitesare green. The former contain more carbon and less oxygen thanthe latter, The classification of graphites as intumescent and non-intumescent fails, because the artificial non-intumescent graphitesgive graphitic acids which are similar to those given by the intu-mescent natural graphites.Cast-iron graphite is an exception,since i t approximates in its behaviour to the natnral non-intumes-cent varieties. Light affects the graphitic acids, causing theircolour to diminish in intensity. I f washed for a long time withwater, these acids are converted into a colloidal modification.The formation of aluminium carbide from its elements has beenobserved a t 750° and 9000, and it has been found that tlie com-~44 W. G. Polack, Tmns. Paraday Soc., 1918, 10, 177 ; A . , 1915, 11, 567.85 K. Arndt, Zeitsch. Elektrochern., 191 6, 22, 63 ; A . , ii, 429.86 A. Lang, Montan. Rundschau,-1916, 19, 1 ; -4., ii, 561lNORGdNIC CHEMISTRY. 59pound is exothermic.87 It dissociates, however, t o a certain extenta t 540°, since when heated in air it forms aluminium oxide andcarbon dioxi4e.Nickel carbide, Ni,C, is an endothermic com-pound, and the optimum temperature f o r its formation is about2100". It dissociates relatively rapidly a t 1600O and more slowlya t 900°. It must therefore be cooled rapidly if a good yield isrequired. Some evidence also was obtained of the direct combina-tion of copper and carbon to give a carbide which is endothermicand dissociate's rapidly a t 1600°, and more slowly a t lower tempexa-tures.Reference was made in last year's Report to the preparation ofperceric potassium carbonate, Ce20,(C0,),,4K2C03,1 2H20.s8 Somefurther viork on these iines may be recorded.89 It was not foundpossible to prepare the corresponding sodium salt by the methodused for the potassium salt, but a solution of the correspondingammonium salt was obtained in that way.If to this dark redsolution e'xcess of solid sodium carbonate is added and the solutionevaporated in a vacuum over sulphuric acid and solid potassiumhydroxide, crystals of perceric sodium carbonate are formed. Thecrystals are much less soluble in water than those of the potassiumsalt, and they eflloresce in dry air. They have the formulaCe20,(CO3),,4N~CO,,30H2O. An interesting analogy between thesodium and potassium salts is shown by writing the formuh asfollows :Ce204 (co3)2,K2co3,3 (K2C03,4H@),Ce~0,(C03)2,Na2C03,3(Na2C03,10H,0).As regards the ammonium salt, although a solution of it has beenprepared, it has not been found possible t o obtain the salt in thesolid state.The corresponding rubidium salt has been preparedby the same method as used for the1 potassium salt, but it has notyet been analysed.Group V .A convenient laboratory method has been described f o r the pre-paration of metallic vanadium, which is suitable also as a lecturedemonstration.90 Vanadyl chloride is prepared by heating a mix-ture of vanadium pentoxidel and charcoal in a stream of chlorine,there being no need to purify i t from any vanadium tetrachlorideit may contain. Some of the vanadyl chloride is placed in a flaskE. Briner and R. Sendet, J . Chim. p h p . , 1915, 13, 351 ; A., ii, 105.C. C . Meloche, J . Amer. Chern. SOC., 1915, 37, 2338; A., 1915, ii, 776.8 9 Ibid., 2645 ; A., ii, 101.90 R.Edson and D. McIntosh, !L'rans. Roy. SOC. Canada, 1915, [iii], 9, 81 ;A., ii, 14360 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.fitted with a platinum wire filament, which may be electricallyheated. The flask should also be fitted with inlet and exit tubesfor the passage of a current of dry hydrogen. The experimentshould be carried out in an atmosphere ol hydrogen a t low pres-sure or in a vacuum. The platinum wire is raised to a white heat,and the metallic vanadium is deposited smoothly as a silvery-greycoating.The crude sodium uranate obtained as a by-product in the ex-traction of radium from carnotite ores contains generally from5 to 10 per cent. of vanadium pentoxide. !L"hree methods aredescribed, by means of which the vanadium can directly be sepa-rated from the sodium uranatel.91 By the action of hydrogenchloride the vanadium can completely bO volatilised, leaving ;tresidue of sodium uranate, sodium chloride, and uranyl chloride.From this residue, 59-64 per cent.of the total uranium can berecovered as the pure oxide, either by boiling with excess ofammonium chloride or by dissolving in dilute acid and precipitat-ing with ammonia, and igniting the ammonium uraiiate produced.I n all probability, by the proper regulation of temperature, hydro-gen chloride would directly effect the quantitative removal ofvanadium from the carnotite ores. Again, by heating the crudesodium uranate with ammonium chloride and sufficient water toform a paste, the vanadium content can be reduced to 0.5 per cent.,and the uranium partly converted into oxide.The best recoveryor" uranium is obtained a t a temperature not exceeding that neces-sary t o volatilise the ammonium chloride and the vanadium com-pound. Finally, by dissolving the crude sodium uranate in theleast possible quantity of hydrochloric acid, or nitric acid, or, insome cases, sulphuric acid, and boiling the solution, the vanadiumis entirely precipitated, together with about 13 per cent, of theuranium. Pure uranium oxide is obtained by adding ammoniat o the filtrate and igniting the ammonium uranate produced. From58 to 79 per cent. of the total uranium can be recovered in thisway when dilute hydrochloric acid or nitric acid is employed.The method of estimation of azoimide in neutral or acetic acidsolution by the measurement of the nitrogen evolved is well known,and it was noted by Rascliig that the reaction is brisk and com-plete when the solution containing the azoimide is mixed with aslight excess of iodine and a crystal of sodium thiosulphate isadded.02 The view has been expressed t h a t the action of the sodiums1 TI.H. Barker and H. Schlundt, Met. and Chem. Erq., 1916, 14, 18 ;g* F. Raschig, Chena. Zeit., 1908, 32, 1203.A., ii, 189INORGANIC CHEMISTRY. 61thiosulphate is due to the catalytic effect of the sodium tetra-thionate, but this is not a possible explanation,93 since neitheriodine alone nor a mixture of iodine and sodium tetrathionate hasany action on azoimide. The suggestion is now made that thereaction between iodine and sodium thiosulphate really takesplace in two stages, the first being Na,S,03 + I, = NaI + NaIS,O,,and it is this intermediate compound that gives up its iodine toanother molecule of sodium thiosulphate.The actual catalyst inthe azoimide reaction would therefore seem to be the residue,NaS,O, which takes up iodine and gives it to the azoimide orcondenses to form Na2S,0,. Since mineral acids hinder the reac-tion, it is advisable to add sodium acetate, owing to the fact thatsulphurio acid is liberated as a by-product. Similarly, it isadvisable from time t o time t o add a crystal of sodium thiosulphateto the solution. It is noteworthy that bromine and sodium thio-sulphate have the same action on azoimide.Although this method is accurate for relatively concentratedsolutions, it does not appear to be suitable for very dilute solu-tions, as the end-point is reached too slowly.94 Salts of quadri-valent cerium, such as, for example, ceric ammonium nitrate orceric sulphate, may be used, for they immediately effect the com-plete oxidation of the azoimide even in very dilute solution.Thereaction takes place according to the equation ZN3H + 2Ce0, =3N2+Ce203+H20. For 0.1 gram of sodium azide about 2.3 gramsof the cerium compound should be used. Free hydrochloric acidor excess of chlorine ions must be avoided, as otherwise chlorine gaswill be liberated.The reachion between azoimide and nitrous acid, which Thieleshowed to take place according to the equation N,H+HNO,=N,+N,O+H,O, has now b'een proved to be quantitative.Theproof corsisted in thO addition of acetic acid to a mixture ofequivalent quantities of sodium azide and barium nitrite, and inthe analysis of the gases involved. The reaction can be used toestimate simple nitritea. A known excess of sodium azide is addedto an acidified solution of the nitrite, then the mixture is shakenfor a minute or two, rendered just alkaline by barium hydroxide,and boiled to expel the nitrous oxide, and finally acidified withacetic acid, when the excess of azoimide is estimated with iodineand sodium thiosulphate. Since nitric acid is not affected, thereaction may be used to remove nitrous acid from a mixture of thetwo.When iodine is added t o antimony pentachloride, reaction takessa F.Raachig, Ber., 1915, 4$, 2088; A., 1916, ii, 98.BI F. Sommer and €1. Pincas, ibid., 1916, 49, 1863; A., ii, 9762 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.place, but bromine, on the other hand, has no action. Three mainreactions take place according t o the equations 95 :SbC1, + I, = SbC1, + 2IC1.2SbC1, + I, = SbC15,21C1 + SbC1,.3SbC1, + 21z = SbC15,31C1 + 2SbC1, + ICl.When less than 1.5 per cent, of iodine is dissolved in antimonypentachloride, the first reaction takes place, and there is no evi-dence for the existence of SbC1,I analogous t o SbF51. The com-pounds SbC15,21C1 and SbC15,31C1 may be obtained by sublimationfrom a mixture of 10 grams of antimony pentachloride with 4.3 or8.6 grams of iodine under 15 mm. pressure at 30-35O.They formbluish-black crystals melting at 62-63O, which fume in air, arereadily soluble in carbon tetrachloride or chloroform, and sparinglyso in antimony pentachloride. The solutions are strongly disso-ciated.Nitrosulphonic acid, HO*SO,-NO,, sometimes reacts as if i texists in the isomeric form, HO*SO,*O*NO, nitrosylsulphuric acid.This suggests that both exist as tautomeric forms in concentratedsulphuric acid solution. It has been found that, if dimethyl-aniline in concentrated sulphuric acid solution is treated with therequisite amount of sodium nitrite, both p-nitro- and pnitroso-dimethylaniline are simultaneously formed.96 This would seem t oprove that the two forms of the acid exist together.Further, a t10-15O the yields of nitro- and nitroso-compounds arel 8.33 and71.45 per cent. respectively, and a t 28-30° they are 42.85 and39-33 per cent., which suggests that the nitro-form is favoured a thigher temperatures.Group VI.It has usually been believed that the action of mercuric chloridcon thiosulphates and polythionates is an oxidising one, owing tothe fact that the white precipitate formed has been assumed to bemercurous chloride. The reaction was supposed t o take place witha trithi,onate, for example, according to the equation:Na2S,0, + 2HgC1, + 2H,O = Na2S0, + Hg,Cl, + 2HC1+ H2S0, + S.Further investigation has shown that the precipitate has the com-position Hg3S2C1,, due t o the combination of mercuric sulphidewith excess of mercuric chloside.97 Thet reactions that take placeare shown by the following equations:2N%S203 + 3HgC1, + 2H20 = Hg,S,Cl, + 4NaCl+ 2H2S0,.2K2S306 + SHgCl, + 4H,O = Hg3S2C12 + 4KC1+ 4H,SO,.2K2S406 + 3HgC1, + 4H20 = Hg3S2C12 + 4KC1+ 4H2S04 + 2s.95 0.Ruff, BeT., 1915, 48, 2068 ; A., ii, 144.J. Biehringer and W. Borsum, ibid., 1916, 49, 1402 ; -4., ii, 560.97 A. Sander, Zeitsch. angew. Chem., 1916, 29, i, ii, 16 ; A., ii, 136INORGANIC CHEMISTRY. 63The validity of these equations has been proved by weighing theamount of the precipitate formed in each case. The estimationof a mixture of trithionate and tetrathionate can be carried outas follows: The solution is oxidised with hydrogen peroxide inpresence of a known excess of sodium hydroxide.The reactionswhich take place are:Na,S,O, + 4H,O, + 4NaOH = 3Na2S0, + 6H20.Na,S,06 + ?H,O, + 6NaOH = 4Na,SO, + 10H20.After the oxidation is complete the unused alkali is titrated.Anot'her sample of the mixture is heated in solution with mercuricchloride, and the acid liberated is estimated. From these tworesults it is easy to calculate the amounts of trithionate and t e t mthionate present.The conclusion that the formula of chromic acid is H2Cr,0, wasarrived a t by Ostwald from his observations on the freezing pointsof aqueous solutions of chromium trioxide, the colorimetric studyof solutions of chromium trioxide, chromates, and dichrornates, andfrom measurements of the electrical conductivity of these solu-tions. Some more recenb spectrophotometric observations confirmthis view.On the other hand, Walden, and also Abegg and Cox,favoured the formula H2Cr0,. Some new measurements have beenmade of the molecular solution volumes and molecular refractivi-ties of chromic acid, potassium chromate, and potassium di-chromate.98 The results of these measurements leave little doubtthat the formula of chromic acid is H,Cr,O,.Mention may be made of the preparation of an interesting com-pound, namely, the methyl ether of the trus orthotelluric acid,Te(OCH,),.Qg The compound was prepared by adding finely pow-dered telluric acid t o a solution of diazomethane in absolute ether.It forms white crystals, which melt a t 85-87O to a limpid liquid.It is soluble in all the usual organic solvents, and, when its aqueoussolution is warmed, i t is partly volatile with the steam, and developsa very irritating odour.Group V I I .Recently, a determination was made of the density of chlorine,and the value obtained for a normal litre of the gas was 3.214grams, from which the atomic weight of chlorine is calculated tobe 35'28.1 This value is considerably below t h a t obtained byg8 A.K. Datta and N. Dhar, J. Arner. Chern. Soc., 1916, 38, 1303; A , ,g B G. Pellini, Gazzetta, 1916, 46, ii, 247.ii, 484.A. Jaquerod and M. Tourpaian, J. Chim. phys., 1913, 11, 3, 269; A , ,1913, ii, 401, 77264 ANKUAL REPORTS ON THE PROGRESS OF CHEMISTRY.chemical methods, and it was suggested that the divergence might,possibly be due to a partial dissociation of chlorine a t the ordinarytemperature.I n order t o test this hypothesis, an investigationhas been made of the vapour pressure of liquid chlorine, of thedensities of liquid and saturated vapour, and of the critical con-stants with the view of determining whether chlorine behaves asa normal substance.2 The chlorine was prepared by heating auricchloride, the greatest possible care being taken as regards itspurity. The vapour-pressure measurements extended from - 78'9Oup t o the critical temperature (144.0), and these can be repre-sented by the formula log P = A - B / T - C log T, in whichA =4-922232, log B=2*9676491, log C= 1.8967405, when P is ex-pressed in atmospheres. The critical pressure was found to be76.1 atmospheres and the critical density 0.573.The densities of liquid chlorine may be represented by theempirical formula d = a + b(144 - t ) + c J 144 - t , in whicha=0'687014, b =O-0002379, and c=0'0622109.The densities ofthe saturated vapour may also be expressed by the same formula,the values of the constants then being a=0.48219, b =0.002451,and c=Om068526. The heats of vaporisation of liquid chlorine a tvarious temperatures have been calculated from the vapour-pressuredata. The ratio of the critical density to that calculated fromthe simple gas laws is 3.635, and the value of Trouton's constantis 20.67. It would thus seem that chlorine behaves as a normalgas, but, a t the same time, further research would seem to beadvisable with regard to the low value for the weight of 1 litreof chlorine referred t o above.The degree of dissociation of bromine vapour into atoms hasbeen determined up to temperatures in the region of 1300O bymeasuring the pressure exerted by a known weight of brominevapour a t a series of temperatures.3 It was found that a t thetemperature of 1284O and the pressure of 721 mm., bromine isdissociated to the extent of 18.3 per cent.Mention may be made of the preparation of various chloritesand a study of certain of their physical properties.4 Certainanalogies between the chlorites and nitrites are described, such as,for example, the existence of the compounds AgClO,,NH,AgC1O2,2NH,, and AgC102,3NH,, which are similar to thoseformed from silver nitrite and ammonia.By the action of a mixture of chlorine dioxide and carbona M. Pellston, J . Chim. phys., 1915, 13, 426 ; -4., ii, 245.M. Rodenstein and F. Crsmer, Zeitsch. Elektrochem., 1916, 22, 337;G. Bruni and G. Levi, Gazzetta, 1915, 43, ii, 161 ; A., ii, 27.A., ii, 552INORGANIC CHEMISTRY. 65dioxide, free from chlorine, on barium peroxide suspended inhydrogen peroxide solution, barium chlorite is obtained absolutelyfree from chloride. The sodium salt may be prepared from bariumchlorite and sodium sulphate, the liquid being evaporated in avacuum a t the ordinary temperature, but the ammonium andhydroxylamine salts cannot be obtained in this way. Among thereactions described for soluble chlorites, the following may bementioned. Potassium f errocyanide is oxidised in acid solution tof erricyanide. With concentrated sulphuric acid, the solid chloritesdeflagrate more energetically than the chlorates. Electrical con-ductivity measurements of potassium, silver, and barium chloritesgave for the mobility of the ion (21021 a t 2 5 O the mean value 51.0,which is less than that for the ion ClO,', and less still than thatfor ClO,'. Similar measurements with sodium, potassium, andsilver nitrites give the mean value 75.4 for the mobility of the ionNO,' a t 25O, this being greater than that for the ion NO,'.It is shown that in compounds containing chlorine in differentstates of oxidation, the formation is the less endothermic or themore exothermic the higher the degree of oxidation.It has been found that the action of chloric acid on iodine ionsin an acid solution, resulting in t.he deposition of iodine accord-ing t o the equation 6H * + 61' + C10,' = 3H,O + (21' + 31, is greatlyaccelerated by the presence of ferri-ions in small concentrations.5The ferri-ions are converted into ferro-ions according to the equa-tion Fe"' + I/= Fe" + I, the ferro-ions being then oxidised thus :6Fe" + 6H' + ClO,' + 3H20 + C1' + 6Fe"'. The method is applic-able to the separation of iodine, not merely from solutions of pureiodides, but from the brine obtained by leaching the ashes of sea-weed with water. To the iodide solution, acidified with excess ofsulphuric or hydrochloric acid, are added (1) potassium chlorate inthe proportion indicated by the above equation, or a greater pro-portion if other substances capable of oxidation are present, and(2) a small volume of a solution of a ferric or ferrous salt in threetimes its weight of water.E. C. C. BALY.L. Pisarshevski and N. Averkiev, J . Russ. Phys. Chem. SOC., 1915, 47,2057 ; A., ii, 184.REP.-VOL. XIII.

ISSN:0365-6217    

DOI:10.1039/AR9161300031

出版商:RSC    

年代:1916

数据来源: RSC

摘要:

ORGANIC CHEMISTRY.PART I .-A L I PHAT I c D I v I s I ON.IN future years it is probable that the work of reviewers in thissection will be modified, and in some measure simplified, by thefact that a sister society has adopted the system of compiling annualrecords of progress in Applied Chemistry. This new developmentis to be welcomed, provided the insidious danger is avoided ofmaking the one report purely theoretical and the other entirelytechnical in its treatment, and although each should make a specialappeal to its own particular audience i t is to be hoped that bothwill be based on the principle that the theoretical expansion of thesubject constitutes the surest foundation for its future applications.These parallel Reports may in time be regarded as a reflectionof the work progreasing in the laboratory as distinguished fromthe factory, and may afford some index of how far these institu-tions are mutually supporting.To the reflective teacher they willalso furnish a test of the efficiency, both in breadth and detail, ofour scientific education.I n preparing the present Report the reviewer has been temptedto take shelter behind the sympathetic reference made recently inthe Journal of the American Chemical Society, where it was statedthat lack of perspective was an inevitable difficulty in reviewingresearch work annually, and that the natural tendency of thereviewers t o devote special attention to the subjects in which theyare particularly interested is by no means to be deplored. I n thecompilation of the present section of the Report very much thesame difficdties were experienced as were referred to last year, andalthough so far as experimental work is concerned the publicationsreviewed under this heading have not been specially shiking, it isa comforting fact that research has been by no means a t a stand-still.Curiously enough, considering the conditions which prevail,theoretical subjects have been prominent, and i t would almostappear as if the partial cessation of orthodox laboratory work hadprovided an opportunity t o take wider and more coinpsehensiveviews. It is to be hoped that some share of this spirit will beGORGANIC CHEMISTRY.apparent in the following pages, as an attempt has been made tomake the Report more than a record of isolated experimental facts.Hydrocarb om.Under this heading the most outstanding feature of the year’spublications has been the appearance of a remarkable series ofpapers dealing with the hydrocarbons involved in caoutchouc syn-thesis.Considerable space must be reserved for the discussion ofthis work, and conseque:ltly references to the simple hydrocarbonscan only be brief.With regard to saturated compounds, there is practically nothingto report, with the exception of the fact that Gomberg has nowconfirmed and extended his earlier observation that among thesimpler reactions of paraffins must be included their capacity toform additive compounds with triphenylmethyl.1 Viewed as anisolated observation, this result possesses little significance, but thequestion of structure a t once arises, and the mod0 of attachmentof the paraffin molecule to animolecular triphenylmethyl becomesa matter of some importance.As has recently been the case, however, unsaturated compoundshave claimed most attention, and the many workers who have hadoccasion to prepare ethylene in quantity, and have observed thatliquid hydrocarbons are formed to some extent during the process,will be intereBted in the results obtained in a detailed examinationof this oil.2 Judging from some earlier observations of Ipatiev, onemight naturally suppose that these hydrocarbons owed their originto simple polymerisation of ethylene, but further examination ofthe mixture has resulted in the separation of a large number ofcompounds in which alkylated polymethylenes predominate, a resultwhich is highly significant in view of recent studies on the auto-condensation of di-olefines. The catalytic hydrogenation ofethylene has also been studied under varying conditions, but theonly point of interest which has emerged is the fact that furtherevidence has been accumulated showing that a factor of primaryimportance in such reactions is the distribution of the catalyst ona supporting medium.Thus, in some cases3 it has been foundadvantageous to use infusorial earth impregnated with metallicnickel, and t o pass the mixture of ethylene and hydrogen throughmolten paraffin. iii which this intimate mixture was suspended.M. Gornberg and C. S. Schoepfle, J .Amer. Chem. SOC., 1915,37, 2569 ; A.,i, 28.G. de MontmoIlin, Bull. SOC. chim., 1916, [iv], 19, 242 ; A., i, 625.J. B. Rather and E. E. Reid, J . Amer. Chem. SOC., 1915, 37, 2115 ; A.,1915, i, 933.0 68 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Many other substances, in addition to infusorial earth, are findingapplication as supporting media in catalytic hydrogenation, but,considering their nature, it is more than probable that the increasedefficiency to which they give rise is due to joint catalytic effects.*Before leaving this subject, reference may again be made to themarked influence of what are apparently insignificant changes inprocedure on catalytic reactions. Thus, in the case of the additmionof acetic acid to acetylene so as t o form ethylidene diacetate, ithas been shown that the reaction proceeds much more smoothlyby forming the catalyst, which in this case consists of mercuricsulphate, in contact with the reacting materials.5 It may be noted,however, that in the closely related case in which water is added toacetylene with the formation of acetaldehyde, the best results areobtained by varying the nature of the solveiit rather than by altera-tions in the method of producing the catalyst.6 These are not unex-pected results, but they serve t o emphasise the difficulty of identi-fying the real functional catalyst in any reaction.Changes which serve as connecting links between the aliphaticand aromatic series are always worthy of attention, even when themechanism of the processes remains obscure. The result of experi-ments in which paraffin-wax has been subjected t o “cracking,”under conditions which are not unduly drastic, seems to show thatdefinite aromatic hydrocarbons are formed even in the absence, inthe paraffin used, of constituents containing phenyl groups.It isdifficult, however, taking into account questions of demand andprobable costs, t o agree with the claim made in the paper now underre vie^,^ that this process may in time form the basis of a methodfor producing benzene and toluene on the technical scale.Caou t c h OIL c.As ahead-y indicated, a large number of publications dealingwith the chemistry of caoutchouc demand detailed consideration.The papers referred t o are concerned with methods of preparingdoubly unsaturated hydrocarbons, the polymerisation of these sub-stances, and the possible structural formulae which should be allo-cated t o the, different forms of caoutchouc. To give an idea of thecomprehensive nature of this work, i t is perhaps sufficient t o statethat in one paper8 no fewer than twenty-nine methods for thepreparation of butadiene are described.These methods naturallyinvolve some processes already known, such as the depolymerisation4 C. Kelber, Ber., 1916, 49, 55 ; A., ii, 309. FT. Pat. 475853 ; A., i, 197.Eng. Pat., 1915, 5132 ; A., i, 465.G. Egloff and T. J. Twomey, J. Physical Chem., 1916, 20, 515 ; A., i, 553.8 I. I. Ostromisslenski, J. Russ. Phys. Chern. SOC., 1915, 47, 1472; A., i, 2ORGANIC CHEMISTRY.69of hydrocarbons allied t o the terpenes, but a considerable numberdisplay much ingenuity, and the possible sources of butadiene havebeen thoroughly explored. Apart from their intrinsic importance,the reactions give an idea of the control exercised by the syntheti-cal organic chemist in removing selected groups from compoundso€ diverse type. For example, the methods involve the eliminationof water, halogen hydride, hydrocarbons, and acids, from open-chain or cyclic alcohols, ethers, halogen derivatives, and esters. Itmay be noted that three of the suggested processes start directlyor- indirectly from alcohol. To take one case which may in timeprove t o be important, the mixture of acetaldehyde and alcoholobtained by the partial oxidation of the latter by the agency ofcopper, undergoes further oxidation when the mixed vapours arepassed over heated alumina, butadiene being then formed.Thechange is by no means so simple as is suggested by the equationCH,*CHO + CH,*CH,*OH -+ CH,:CH*CH:CH, + 2H,O, but in-volves no fewer than five consecutive reactions.9 Still, the possi-bility of commencing one type of caoutchouc synthesis from alcoholis attractive, particularly as the method seems to be genera1,lOand even admitting that the yield obtained by the above method ispoor, this is a matter which will probably be improved.The goal of Ostromisslenski’s work is apparently the production,by synthetical means, of elastic colloids which display the essentialphysical characteristics of natural caoutchouc.With chemicalstandards of comparison between synthetical and naturaj productshe is apparently little concerned. Some idea of his position isgleaned by noting his divergence from Harries in the view as towhat should be included under the expressions “normal” and‘‘ abnormal” caontchoucs.1l The formation of ozonides, and theidentification of the decomposition products to which they give rise,have in the past been made the basis of differentiation between“ normal ” and ‘ I abnormal” types, but the, view now advocated isthat a caoutchouc should be classified according to the temperaturma t which the colloid acquires and loses its elastic properties, andt o the range of temperature between these points. When thesefactors agree approximately with those determined for naturalcaoutchouc, the substance under examination in termed “ normal.”Although this may appear empirical, there is much to be said f o rsuch a view, and the effect of its adoption would be to relegatecaoutchouc, and structurally related substances, t o a single divisionof a large class of colloids which show similar physical properties.9 I.I. Ostromisslenski, J . Russ, Phys. Chem. SOC., 1915,47, 1494 ; A . , i, 4.1 0 I. I. Ostromisslenski and P. N. Rabinovitsch, ibid., 1507 ; A . , i, 4.11 I. I. Ostromisslenski, ibid., 1374 ; A., i, 5470 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Considering the conflicting results which have been obtained bydifferent experimenters in decomposing caoutchouc ozonides, thechemical method of classification does not seem in the meantimet o be any more definite.Despite his views on classification, Ostromisslenski does not hesi-t a t e to extend his work from the synthesis of the parent hydro-carbons to their polymerisation to caoutchouc, and the methodsdescribed by him for these conversions are apparently productiveof good yields, although they add some complications t o questionsof structure.Thus, to take one case,12 isoprene has been shown t oundergo auto-condensation, when preserved a t 80-90°, t o give anopen-chain dimeride termed P-myrcene. This compound containsthree double linkings, two of which are in conjugated positions,and, when polymerised with barium peroxide and sodium, yields anormal caoutchouc.Curiously enough, direct polymerisation ofisoprene by the same agency gives an abnormal product, the outlineof the scheme being as follows:2nC,H, + abnormal caoutchoucJ.nC,,H,, + normal caoutchoucObviously, t h e next step required to throw light on the poly-merisation process was to place the! structure of P-myrcene on asatisfactory basis, and to gain some' idea of the magnitude of thefactor n.Although his views are largely speculative, Ostromisslenski hasa strong claim, in view of 'his successful work, t o be heard on thesubject of the structure of caoutchouc, and the mechanism of itsformation. Basing his arguments13 largely on the fact t h a t thebromide of natural caoutchouc is homologous with caoupreiiebromide, and is thus unicyclic, he comes t o the opinion t h a t caout-chouc itself must similarly be unicyclic, the ring containing thegroup (*CR,*CH,*CMe:CH*),.I n this respect he does not differfrom the latest conclusion arrived a t by Harries. Accepting thisprinciple,14 we are faced with the view that., if the caoutchoucs areunicyclic, the, individual molecules must consist of unusually largerings, and by its application to the formation of P-myrcene15 andthe subsequent polymerisation, a fairly clear account of the forma-tion of normal isoprene caoutchouc can be constructed. As aninitial step, isoprene may be regarded as undergoing a species of12 I. I. Ostromisslenski and F. F. Koschelev, J . Ritss. Phys. Chent. Soc., 1915,l3 I. I.Ostromisslenski, ibid., 1932 ; A., i, 274.I. I. Ost,romisslenski, ibid., 1937 ; A., i, 275.l5 I. I. Ostromisslenski, ibid., 1941 ; A . , i, 276.47,1928 ; A., i, 274ORGANIC CHEMISTRY. 71dissociation into hydrogen and a residual radicle, in whicli case theformation of the dimeride will follow the route shown below:C€I,:C'RIe*C'H:CH, + H + CH:CMe*CH:CH,.Ths unsaturat,ed fragment then combines with a second molecule ofisoprene, the adjustment of the valencies being effected by the newposition assumed by the free hydrogen atom:CH2: CMe*CH2*CH,-CH:CMe CH: CH, (0-myrcene) .Further action involves the dissociation of one of the hydrogenatoms rn-arked x and open-chain polymerisation on lines similar tothe above then proceeds until the octameride is formed, a t whichstage ring formation puts an end to the possibility of further com-plications.It must be admitted t h a t these theoretical ideas are somewhatcrude, and suited only for blackboard demonstration.Consider-ing current opinions as t o the nature of valency and certain valu-able suggestions, to which reference will afterwards be made, ast.2 the general mechanism of addition to unsaturated molecules,the scheme outlined above must be regarded as somewhat super-ficial.Numerous papers on the subject of vulcanisation have also ap-peared from Ostromisslenski's laboratory, and although of specialinterest t o the technical chemist, they contain many importantobservations and suggestions. Harries has also contributed a paperon the same subject, but the somewhat empirical results describedare beyond the scope of the Report.The general impression left in the mind of the reviewer aftera survey of the original work in this field f o r the past five yearsis t h a t although the synthesis of a caoutchouc, identical in chemicalstructure and physical texture with the natural product, may neverbe achieved, it is evident t h a t as a result of this work many syn-thetic substances possessing valuable physical properties will intime be rendered available, and will find extensive application.The development has given considerable stimulus t o theory, andhas revived interest in the study of hydrocarbons generally.I nleaving the subject, passing reference may be made t o an excellentaccount 16 of the past history and present position of the caoutchoucproblem, which will be found useful to place in the hands ofstudents who are making their first acquaintance with the subject.X14i~coho7s C L I ~ t A e i r UerivritivP.9.So fa:.as monohydric aliphatic alcohols are concerned, there isvery littlle of general int'erest to report, most publications under thislG 13. I). W. Luff, J . Xoc. Cheni. Id., 1916,35, 98372 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.heading having been confined t o strictly technical inquiries. In afew cases paDers have dealt with experimental methods of purifyingalcohols, and more than passing interest will be attached t o a briefnote on the precautions to be observed in the dehydration of ethylalcohol by metallic calcium.To those who have in the past usedanhydrous alcohol purified by this agency f o r exact physical work,the observation that minute traces of ammonia are frequently pre-sent will not come as a surprise, and a convenient method17 forremoving this impurity by the use of alizarin and tartaric acid willbe welcomed.The hope expressed in a former Report t h a t the whole subjectof the oxidation of alcohols by catalytic methods would be system-atically studied, has unfortunately not been realised, and the workhas apparently been suspended. This is t o be regretted, as muchremains to be done in this field, and the methods now i n use forthe graded oxidation of even the simplest alcohols on a large scaleare by no means perfect.To take a case in point, the conversionof methyl alcohol into formaldehyde under the influence of heatedcopper is a process which is notorious for the irregular yields andvarying purity of the product obtained. One factor affecting theresult is, of course, the liability of methyl alcohol t o be contamin-ated with acetone, but leaving this out of account, the reaction isapparently by no means so simple as was a t one time believed to bethe case. Even when conducted a t moderate hmperatures, theformaldehyde liberated shows a tendency to undergo degradationto hydrogen and carbon monoxide, but a t the same time, owing topolymerisation, a quantity of methyl formate is produced. Theseresults, which have been confirmed in the course of attempts toapply the oxidation process analytically,l8 have an important bear-ing on some of the difficulties encountered on the technical scale.It has also been noted t h a t the use of this metallic catalyst occa-sionally gives a product contaminated with copper compounds, andthe suggestion has been made t h a t this result is due to the forma-tion of either copper formate or copper carbonyl, both of whichare volatile.19 This conclusion is probably justified, although ithas been shown20 that copper formate is easily decomposed onheating with the formation of formaldehyde.It may be remarkedt h a t methods are now available f o r the removal of copper com-pounds from formaldehyde, but it is evident that, for severalreasons, copper is not an ideal catalyst for the reaction in question.The use of peroxides as oxidising media has recently found fewl7 L.W. Winkler, Zeitsch. angew. Chem., 1916,29, 18 ; A., i, 245.18 C. Mannich and IV. Geilmann, Ber., 1916,49, 585 ; A., i, 362.19 H. Kunz-Krause, Chem. Zentr., 1916, i, 554 ; A . , i, 645.2o K. A. Hofmann and K. Rchumpelt, Ber., 1916,49, 303 ; A . , i, 369ORGANIC CHEMISTRY. 73applications, but, on the other hand, the, application of ozoneto the determination of the position of unsaturated linkings inalcohols continues to yield fruitful results,21 although it is evidentthat, in structural work involving the formation and cleavage ofozonides, due regard must be paid to newly recognised types ofdecomposition undergone by these compounds.22Aldehydes and Ketones.The progress made in recent years in the study of unstablealdehydes has not been maintained in the period now under review.I n fact, most recent work on aldehydes and ketones is somewhatill-defined, and the only results which might be mentioned areextensions, on standard lines, of previous investigations or minorimprovements in working processes.Amongst these may bequoted new examples of the application of Skita's method ofcatalytic reduction as applied to compounds possessing thecarbonyl group, especially as the paper in question contains thecomforting suggestion that the high pressures generally adoptedby Ipatiev in such reactions are not always necessary.23 Anotherexample of improved practical methods is furnished by the observa-tion that aminonitriles can be formed from both aldehydes andketones by a simple modification of the customary proced~re.23~Glacial acetic acid is used as the solvent for both the carbonylcompound and amine, the solution being afterwards mixed withconcentrated aqueous potassium cyanide.Under these conditions,no hydrogen cyanide escapes, and the acetic acid not only promotesthe condensation, but functions as a crystallising medium for theproduct. Suggestions of this nature are by no means trivial, asthey minimise working dangers, and thereby add t o the range ofapplication of a reaction.The constitutional and experimental difficulties surrounding thesynthesis of definite glycerophosphates are well known, and it willbe remembered that an important advance was made two yearsago, when the synthesis of a-glycerylphosphoric acid was described.The method then employed was the interaction of trisodium phos-22 C.Dor6e and L. Orange, T., 1916,109,46 ; A., i, 261.22 C. Harries, Anna.len, 1915,410, 1 ; A., 1915, i, 966.23 A. Skita and P. Stuckart, Ber., 1915,4!8, 1486 ; A., i, 16.23~6 R. von Walther and R. Hiibner, J. pr. Chem., 1916, [ii], 93, 119 ; A.,i, 559.D‘74 ANRUAL REPORTS ON THE PROGliESS OF CHEMISTRY.phate and a-monochlorohydrin; but i t has now been shown thatthe change is by no means simple, and involves the intermediateformation of the glycide, as shown below :0/\OH*CH,*CH(OH)*CH,Cl + OH*CH,*CK*CH, + Na,HPO, --+0 H CH, C H (0 H) C H, 0 P 0 ( 0 N a) ?.It is quite apparent that this detailed study24 of the course ofthe reaction does not invalidate King and Pyman’s work, nor isit likely that the glycerophosphate finally produced is a mixtureof steroisomerides, despite the fact that the opening of the glycidering affords two possibilities for the subsequent addition. A con-siderable amount of direct and collateral evidence supports theview that the phosphate residue is attached in the terminal posi-tion, and it is interesting t9 note that, a somewhat similar sugges-tion has been made by Abderhalden in order to explain thebewildering results obtained by him in his efforts to synthesiseoptically active fats.This particular topic has been the subjectof appreciative references in past Reports, and it is satisfactoryto know that steady progress is being made despite accumulatingdifficulties.25 From one point of view? perhaps the most importantnew result now to hand in this subject is the marked differencewhich exists bet,ween the reactions of the active mono- and di-bromohydrins.I n each case, halogen may be replaced by theamino-group, but whereas an active monobromohydrin gives riseto an active aminoglycerol, similar treatment of an active dibromo-hydrin gives an inactive diaminohydroxypropane. The explana-tion offered is that the above reactions are not, to be regarded assimple exchanges of groups, but involve the intermediate forma-tion of a glycide ring from the monobromohydrin, and the con-secutive production of two such rings in the case of dibromo-hydrin.0/\CH,Br*CHBr*CH2*OH (active) --f CH2Br*CH-CH, +0/\CH,Br=CH(OH)*CH,*NH2 --+ CH,*CH*CH2*NH, -+NH,* CH,*CH( OH) CH,-N H, (inactive).It will be seen that the ultimate product is internally compensated,but that the adoption of a similar structural scheme would not24 0.Bailly, Compt. rend., 1915,161, 677 ; A . , i, 113.25 E. Abderhalden and E. Eichwald, Be?., 1815, 48, 1847 ; .4., i, SORGANIC CHEMISTRY. 75interfere with the production of an active aminoglycerol frommonobromohydrin.2GThe latter type of active compound is not, however, entirelyproof against racemising agencies, and even when brought intoreaction with metallic salts of fatty acids, totally inactive glycerolesters were obtained.MonobromohydrinDibromohydrinMono-estersor (inactive) 1 Di-estei sor ’active + R*CO,M +-It is highly important to note that, once the amino-group Lasbeen introduced into the glycerol molecule, ester formation with-out racemisation is possible by means of the sulphuric acid method,so that definite active glycerol esters have been obtained accord-ing t o the scheme:Glycerol bromohydrin ?!$ aminoglycerol + aminoglycerol esterNa 9 O2 ___, glycerol ester.I n addition to the difficulties outlined above, the work has beencomplicated by curious optical inversions, of which the followingmay be taken as typical:d-monobutyrin.A .1-Epihydrin alcohol 7’ ‘ d-aminoglycerol -+ I-inonobutyrin.HC1 KOH B. d-Epibromohydrin -+. chlorobromopropan-/3-ol -+I-epichloroh ydrin.Considering both types of reaction, it is evident t h a t syntheticalschemes can be constructed resembling Walden inversion cycles,and the conclusion arrived a t in the original paper is that, theactivity of a glycerol ester always varies in sign according as thesubstituting group is introduced in the a- or y-position of theparent epihydrin alcohol.I n all probability, our structural views regarding optically activepolyhydroxy-compounds will be considerably aff ecbed by the studyof derivatives in which selected groups are substituted.Fischerhas recently turned his attention to this problem, and during theyear has contributed two papers,2”2* which contain a number ofinteresting observations. He finds, f o r example, t h a t the mannitol-diacetone obtained by the incomplete condensation of mannitolwith the ketone is isomeric, and not identical, with the mannitol-26 See also K.Hess and H. Fink, Ber., 1915, &, 1986 ; A., i, 158.27 E. Fischer and M. Bergmann, ibid., 1916,49,289 ; A., i, 364.28 E. Fischer and C. Rund, ibid., 88 ; A., i, 363.D* 76 ANNUAT, REPORTS ON 'I'HE PROGRESS OF CHEMWI'RI'.diacetone which results from the partial hydrolysis of the triacetonecompound.Mannitol / Adiaceto; \ II monoacetone ! -+graded condensation ' 7 trincetoneIr monoacetone + diacetoneA B graded hydrolysis.The course of the reaction from A to 13 thus differs in mechanismfrom the reverse process, and this is in strict agreement with theviews expressed by Irvine and Paterson as to the stereochemicalstructure of mannitol.Another significant result is also quoted which shows thatdulcitol offers similar opportunities for stereochemical study, andgives partly substituted derivatives possessing unlooked-for proper-ties.To take a case in point, two dulcitoldiacetones are known(a and a), and the natural assumption would be that in thesecompounds the isopropylidene residues are attached to differentpairs of carbon atoms. The result of benzoylation should there-fore substitute the remaining hydroxyl groups, and thus preservethe isomerism; but such is not the case, as shown in the followingscheme :Dulcitoldulcitol-a-diacetone dulcitol-P-diacetonex \P-dibenzoyldulcito1:a-diacetone a-dibenzoyldulcitol-a-diacetoneIx 4 I \ 4a-dibenzoyldulcitol + a-dibenzoyldulcitol.Not only does acylation obliterate the isomerism between the a-and P-diacetones, but two forms of dibenzoyldulcitol exist, one ofwhich can be converted into the other.From these results i tfollows that in both a- and P-dulcitoldiacetones the same hydroxylgroups must be unsubstituted ; nevertheless, the compounds areisomeric. These facts receive a ready explanation in terms of thetheory that the primary hydroxyl groups in polyhydroxy-corn-pounds, such as dulcitol, are differently arranged in space, and itmay be mentioned that somewhat similar considerations have beeORGANIC CHEMISTRY. 77applied to explain the varying capacity of dibasic acids to formadditive compounds.29Acids and their Derivatives.Comparatively few of the experimental results bearing on thechemistry of the aliphatic acids which have been noted duringthe period under review call for special mention, and thus spacemay be reserved under this heading for reference to importanttheoretical views which have been put forward in the course ofthe year.Meanwhile, although the subject of optical activity hasbeen discussed a t some length ia recent Annual Reports, mentionshould be made of Fischer's renewed attempts to verify by opticalmethods the principle of the inter-equivalence of the four carbonvalencies. The idea underlying the work30 is that, by alteringthe different groups attached to an asymmetric atom so that theybecome identical in pairs, the equivalence of the valencies involvedwill be indicated by the disappearance of optical activity.Con-ditions were naturally selected which would be free from Waldeninversion effects, so far as these perplexing possibilities can bepredicted. Starting from ethyl a-cyanovalerate, the allyl groupwas introduced into the butyl residue, the product thereafter beingconverted into a-cyano-a-allylvaleric acid, which was resolved intoits active forms. Reduction of the acid under mild catalytic con-ditions gave a-cyano-a-propylvaleric acid, which was quite inactive,thus proving the equivalence of the two valencies marked a and bin the formula shown below:Unfortunately, the extension of the scheme t o proving the equi-valence of the valencies c and d failed, owing to the fact that theallyl residue participated in hydrolysis reactions which weredesigned t o affect only the nitrile group.Topics of this natureare certainly interesting, and, as in many problems affecting com-pounds of simple structure, the experimental difficulties seem tohave been considerable, but it may be remarked that structuralevidence based on the disappearance of optical activity must ofnecessity be invested with some uncertainty.Another optical study which may be mentioned is the prepara-tion of the active thiolmalic acids, which have been obtained bytwo alternative methods: (1) the action of aqueous ammonia on2 8 T. 8. Price and 8. A. Brazier, T., 1915,107, 1713 ; A . , i, 131.3" E. k'isclier aiid 717.Brieger, Ber., 1915,48, 1517 ; A,, i, 11'i8 ANNUAL REPORTS ON THE PROGRESS OF CIlEMISTHP.active xanthosuccinic acid, and (2) by the interaction of potassiumhydrogen sulphide and active bromosuccinic acid.31 By alteringthe conditions, the latter reaction could be varied t o give differentmixtures of the d- and I-product, the explanation offered, whichis similar t o that, already put forward in tlie case of malic acid,being that the change may proceed directly or through the inter-mediate formation of the active lactone of malic acid:,P\ \Ci)*CH,*CH*CX~,H.Other reactions of bromosucciiiic acid support in some measureHolmberg's views, but somewhat more complex possibilities areopened out by the results of another research? in which it isshown that the reaction between bromosuccinic acid and pyridinebases is complex and follows an unusual course.From othersources, our knowledge regarding the lactones of the malic acidseries is gradually being extended, particularly with regard to thedecomposition of these compounds when heated, and a recentpaper s3 furnishes an interesting result illustrating how the mole-cular rupture undergone by these compounds is controlled by theposition of the lactonic group. The methyl ester of as-dimethyl-maliclactonic acid, although stable enough t o be distilled at theGaede pump, suffers decomposition when boiled under 18 mm.pressure, giving rise t o carbon dioxide and methyl PP-dimethyl-acrylate. On tlie analogy of previous work, dimethylketen andacetaldehyde might have been expected to be the essential productsof the change, so that the molecular rupturc? evidently follows adifferent course, according t o the position of the substituent alkylgroups, as showii below :O--CR!h, CO- -CMe,1 1 I I1 1 : 0--- -CH-CO,Me CO-CMe- -co,:M~Of the quantitative methods available for the study of theketo-enol equilibrium, i t is difficult to decide which is the mosttruscworthy, and the examination of oxalacetic acid by the spectro-graphic method 34 furnishes another example where results havebeen obtained which are opposed t o those arrived a t by the useof bromine i n this determination. I n the paper referred to, someingenious suggestions are made as to the existence of hydroxy-31 B.Holmberg, Arkiv Kenz. Min. CleoZ., 1916,6, NO. 1, 1 ; A., i, 307.32 0. E. Lutz, J . Rziss. Phys. Chem. SOC., 1915,47,1549 ; A . , i, 73.33 E. Ott, Rer., 1915,443. 1350: A . , 1916, j, 1050.34 A. Hantzsch, ibid., 1407 ; 3., i, 13ORGANIC CHEMISTRY. 79maleic acid as a conjugated cyclic structure, but consideration ofthis idea may be withheld in favour of reference to a mostimportant contribution which has recently appeared on the generalsubject of unsaturation.A considerable expansion of our views owes its inception to theobservation that the presence of tervalent nitrogen in attachmentt o a doubly linked carbon atom gives rise to the properties charac-teristic of a conjugated system. The actual compounds whichfurnished this result fall within the province of another sectionof the Report, but the following skeleton scheme will explainsufficiently clearly the essentials of a typical reaction in illustra-tion :CHPh:C-NMe + HI -+ CH,Ph*C=NMeI.Addition thus results in the formation of a methiodide, and thisinvolves the shift of the double bond from C:C to C:N.In orderto account for this divided addition to what may be regarded asthe conjugated system (ethylenic carbon and tervalent nitrogen),the exercise of partial valency becomes a necessary assumption,and, as there is no good reason why such partial valencies shouldbe restricted to any particular form of attachment, the ultimatedevelopment of this idea is the view that a unit of partial valencyaccompanies each normal valency.Accepting this conclusion, thecarbon and nitrogen atoms may thus be represented as/ \ / \I- ::. . .c.. . . . .....so that, in a sense, each single bond may be regarded as unsatu-rated. It is, however, unnecessary to call into play the full numberof possible partial valencies in order to explain the addition reac-tion now under consideration, as is seen by inspection of thediagrammatic scheme used by the authors.35\\ ,,: ..... yc\\ ,..Yd \Cdeyde deA preliminary union of the reactive molecules by means of partialvalencies, followed by adjustment to the most stable type, com-pletes a very simple explanation of the addition reaction. Appli-cation of the above principle of temporary addition by partial35 Miss E. E. P. Hamilton and R..Robinson, T., 1916,109, 102980 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.valencies is given in the original paper, and of these the moststriking is perhaps the simplification of Stewart’s structural repre-sentation of the maleic + fumaric acid transformation-an ex-ample which, despite the draughtsmansliip required for its repro-duction, will doubtless find its way to the lecture-room. Beforepassing on to the applications of the ethylene-nitrogen conjugationtheory to the aliphatic series, reference may be made t o anotherimportant contribution36 to the study of valency which, by astriking coincidence, appeared in the same issue of the Journal.The subject-matter of the paper in question is quite outside thescope of the present section, but, this fact should not preclude anappreciative reference t o a highly interesting paper which, in viewof the representation of the nitrogen valencies suggested, shouldbe read in conjunction with the two papers now under review.To resume.By application of the theory of nitrogen-carbon con-jugation to ethyl P-aminocrotonate, a new light is thrown on themechanism of the reactions of this compound and its homologues.From the ease with which the ester is hydrolysed by acids, thereasonable conclusion is drawn that the change proceeds throughpreliminary addition to the conjugated system.conjugatedNH,CI + O:CMe=(’IH,=CO,Ft.A similar addition of ethyl iodide, followed by hydrolysis, alsoaccounts readily for the formation of ethyl a-acetylbutyrate fromaminocrotonic ester, and strong support is given to the theoreticalscheme by the reactions of ethyl P-diethyJaminocrotonate,NEt,-CMe:CH*CO,Et. AIthough non-reactive towards sodium,this ester gives additive compounds with methyl o r ethyl iodideswhich, on decomposition with water, yield acetylpropionic andacetylbutyric esters respectively.NEh*CMe:CH=CO,Et + R I + NEt,I:CMe*CHR*CO,Et +What is doubtless the most convincing argument in favour ofthese speculations is, however, furnished by the fact that, by theagency of additive reactions, an alkyl group may be attached to acarbon atom which is not coupled t o hydrogen.The compoundto furnish this important result was dimethyl ethylenebis-p-amino-a-methylcrotonate, which, after addition of methyl iodide andsubsequent hydrolysis, was found to have undergone C-alkylation,despite the absence of the customary reactive hydrogen atom.(C02Me*CMe:CMe*NH*CH2), --+ 2CO,Me*CMe2*COMc.0: CMe-CHR-CO,E t .36 A.Clayton, T., 1916, 109, 104ORGANIC CHEMISTRY-. 81With these convincing results as a foundation, it is but naturalto inquire how far similar ideas extend to the reactions of ethylacetoacetate, and the substitution of ONa for NEt, in the struc-tural scheme given will show their ready application in thisparticular case. The further exploitation of the theory, particu-larly on the lines reserved in the original paper,37 will be awaitedwith much interest.Some topics of research, from the theoretical difficultieswith which they are surrounded and the mass of experimentalwork which they involve, are not readily discussed within thelimits of the Annual Reports.I n such cases, the reviewer has noalternative but to direct attention at once to the original papers,and thus avoid the risk of doing scant justice to closely reasonedand sustained arguments by an imperfect synopsis. This applieswith special force t o the modern type of investigation on opticalactivity, and, in the case of recent communications3* on the sub-ject, little more can be done than t o give a bare outline of thenature of the work. The general principles laid down by Patter-son as to the form of temperature-rotation curves have been sup-ported by the results of a new series of optical determinationsconducted on the alkyl tartrates, both in the homogeneous stateand in solution.I n this way, a large addition has been made tothe available data showing the effect of temperature and of solventson the rotations of these compounds in light of different wave-lengths. An important point which has emerged is that tempera-ture-rotation curves may be extended in either direction beyondthe limits imposed by the boiling point or melting point of thehomogeneous ester by using solutions in selected solvents. Thetheoretical treatment of the results involves a general criticismof much that has recently been published on the causes of abnormalrotatory dispersion, and in particular on the suggestions whichhave been made regarding the possibility of isodynamic change inethyl tartrate.It is extremely important to have this pointsettled and to secure some trustworthy method o l determiningwhen a substance is optically homogeneous, but the present posi-tion of this branch of research, if only for the reason that it isone in which experts differ and have not hesitated t o argue theirpositions, is far from unsatisfactory even if the ultimate goal isstill distant.A considerable number of new esters have been prepared fromaliphatic acids, but, for the most part, these preparations havebeen undertaken for pharmacological purposes and display no37 R. Robinson, T., 1916, 109, 1038.38 T. S. Patterson, ibid., 1139, 1176, 120482 ANNUAL REPORTS ON THE PROGRESS OF CIIEMISTRY.novel features.Again, with regard to reactions of esters, thesehave, almost without exception, been studied on normal lines.Naturally enough, the Grignard reagent continues to be exten-sively applied to these compounds, and attention may be directedto a series of papers by Stadnikov which are important principallyfor the irregularity of the results obtained. Such work is of value,as it serves to throw light on the mechanism of what are usuallyregarded as side-reactions, and their study may possibly have theeffect of making successful use of the Grignard reagent lessdependent on practice and experience than is frequently the case.It may be noted that the magnesium haloid which is always pre-sent in the reagent may interfere with the reaction, owing to itstendency to react with an ester and generate a different organichaloid from that originally introduced.39 The initial step of thistype of side-reaction is indicated below :Incidentally, however, it may be mentioned that some of thevariations encountered by Stadnikov apparently owe their originto the employment of higher temperatures than are usually adoptedin these reactions,40 or may even be attributed to the method ofmixing the reagents.41Although reference has already been made t o new ideas regard-ing the mechanism of syntheses based on ethyl acetoacetate, atten-tion may a t this stage be directed t o some interesting results whichhave been obtained in attempts to synthesise hydrocarbons alliedto the terpenes.42 Condensation of chloroniethgl ether and methylsodio-a-methylacetoacetate gives the enolic product,OMe*CH,*O*CMe:CMe-CO,Me,from which the corresponding methoxy-P-methoxy-a-methylcrotonicacid was obtained.The decomposition of this acid by heat dis-engaged carbon dioxide as usual, but the reactions of the essentialproduct showed that, as the unsaturation was preserved, i t is con-sequently to be regarded as a butylene derivative,OMe* cH,* 0 CMe: CH- CH,.This type of reaction is not confined to one example, and addi-tional evidence as t o the irregularity of the above substitutedcrotonic ester is furnished by the fact that the products of itscondensation with either methyl or ethyl acetoacetates proved tobe complex, and their structure is still undetermined.An observation which is important in that it makes a valuables9 G.L. Stadnikov, J . RZ~SS. Phys. Chem. Xoc., 1915, 47, 1122 ; A . , 1915, i,4 0 Ibid., 2037 ; A., i, 259. 41 Ibid., 2115 ; A . , i, 260.43 A. Lapworth and B. S. Mellor, T., 1915,107, 1373 ; L4., 1015, i, 039.R*CO,R’ + MgI, + R-CO,*MgI + R’I.957ORGANIC CHEMISTRP. 83reagent accessible is furnished by the discovery that aliphaticnitriles can be conveniently prepared by a simple modification ofLett’s process.43 I n the particular case of acetonitrile, a catalyticmethod of preparation is also described 44 which is continuous inits action and gives excellent yields. The process depends on thedehydration of acetic acid and ammonia by passing the mixedvapours over alumina o r thoria heated to 500°, and the successof the method suggests that it may be capable of extension to thepreparation of nitriles from any saturated acid possessing therequisite volatility.Curb oh ydra t es.Recent work on the sugars furnishes a striking instance of therenewal of interest in a group of compounds, the study of whichwas apparently on the decline.Subsequent to the time whenFischer’s synthetical work in this field diminished from its firstactivity, there succeeded a period which naturally suffered bycomparison with the brilliancy of that which preceded it. Duringthis interval, however, much steady and valuable progress wasmade which has prepared the way for the present revival and forfuture advances, which promise to be of great importance.During the past three or four years, the pages of the variousjournals bear ample testimony to the fact that research on thesimple sugars is now attracting many workers (both old and new),and is showing a distinct tendency to expand.Questions of con-stitution are still prominent, and, as a necessary result, syntheticalwork is receiving a new impetus. Speaking generally, the experi-mental study of sugars may, a t the present time, be classifiedunder four heads: (1) investigations on the structure of reducingsugars; (2) the synthesis of partially substituted sugars; (3) therelationship between rotatory power and configuration ; (4) syn-theses effected by enzyme action. All these developments areaffected by the recognition now given to the existence of reducingsugars in forms other than the recognised a- and P-types, and thisconsideration should be kept strictly in view both in reviewingold observations and in the interpretation of all new results.It has been noted more than once in the Annual Reports thatalthough most of the reducing sugars have been isolated in well-defined, mutarotatory forms which possess, presumably, they-oxidic linking, other isomerides may exist and, in certain cases,do actually exist.Thus, taking glucose as an example, thestandard crystalline forms of the sugar (a and /3) correspond with43 G. D. van Epps and E. E. Reid, J . Arne?.. Chem. Soc., 1916,38, 2120.44 I b i d . , 212884 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.the isomeric crystalline methylglucosides.I n addition, however,two other methylglucosides have been proved to be present in theso-called (( y-methylglucoside,” but so far no crystalline variety ofglucose has been isolated other than the above a- and P-forms,either individually or as mixtures. The existence of no fewerthan four methylglucosides shows that the parent sugar must reactin four forms, in two of which the ring-forming oxygen atom doesnot occupy the customary y-position.Considering the cyclic structures possible in a hexose, it isevident that, as a maximum, there may be five isomeric glucosesin addition to the aldehydic type, each capable of existing in a-and j3-forms, and similar considerations may well apply to allreducing sugars. Little systematic progress has been made duringthe past year in the important problems thus opened out, but aconsiderable amount of indirect evidence has been obtained whichsupports the views expressed above.Thus, Hudson and his co-workers, in the course of ’their investigations on rotatory powerin the sugar group, isolated a second form of galactose penta-a~etate,~5 and followed this up by the preparation of a third46and a fourth variety.47 According t o accepted views of structure,galactose should only give two non-reducing penta-acetates, butthe isolation of four such compounds shows that this sugar mustexist in at least four forms, and the satisfaction of having isolatedthe first series of crystalline derivatives corresponding with thenewly recognised type of sugar isomerism thus rests with the Ameri-can School.Close inspection of some of Fischer’s recent work likewise revealsevidence pointing similarly to the idea that we are only a t thebeginning in unravelling the complexity of even the simplestsugar molecules.Although Fischer himself is meanwhile of adifferent opinion, good reasons exist for the belief that the com-pound known as glucosemonoacetone is derived from the hypo-thetical “y-glucose,” and in such case it would follow that thetrisubstituted sugars obtained from glucosemonoacetone would alsobelong t o the y-type. Trimethyl glucose was the first repre-sentative of this class, and the corresponding tribenzoyl glucosehas now been isolated as the result of similar processes.48 Thenew sugar is a viscous oil and displays well-marked additive proper-ties, these features being characteristic of members of the 7-glucoss45 C .S. Hudson and H. 0. Parker, J . Amer. Chem. Soc., 1915,37, 1589 ; A . ,46 C. S. Hudson, ibid., 1591; A., 1915, i, 651.47 C. S. Hudson and J. M. Johnson, ibid., 1916, 38, 1323 ; A . , i, 546.48 E. Fischer and C. Rund, Zoc. cit..1915, i, 652ORGANIC CHEMISTRY. 85series, so far as these compounds have been examined. In thecourse of the research now under discussion, a description is givenof an improved method of preparing glucosemonoacehne bycautious half-hydrolysis of glucosediacetone, but it may beremarked that the details of precisely the same method were pub-lished more than a year ag0,4Q and were mentioned in last year’sReport.I n view of the difficulties which will have to be faced in assign-ing a structure to the newly recognised varieties of sugars, it issomewhat disconcerting to find that current ideas as to theisomerism displayed by a- and /3-glucose are still the object ofcriticism.60 The arguments put forward by Anderson in supportof the view that a- and P-glucose do not differ stereochemically, butin respect of the nature of the internal ring present in each mole-cule, are certainly worthy of consideration, but are opposed tomuch clear-cut experimental evidence t o the contrary, and maythus, meanwhile, be held in reserve.Another paper 61 involvingstructural considerations to which attention should be directed con-tains a number of valuable experimental results bearing on theconfiguration of sugars containing more than six carbon atoms.The work in question is undoubtedly of a high order, but in clear-ing up some points has had the effect of adding confusion toothers.This is not so much the fault of the investigator as ofthe totally ingdequate nomenclature with which the chemist inthis field is equipped, or rather handicapped. As a general rule,the first of several possible isomeric sugar derivatives isolated ispromptly termed the a-form. If, in the course of time, a secondand third isomeride are obtained, these become known as the P-and y-forms, irrespective of the fact that the index letters a and /3refer t o definite stereochemical isomerides and have a restrictedapplication.Thus, Pierce obtained a new variety of mannoheptoicacid, which is termed the /3-form, for no better reason than thatit is the second such compound t o be isolated. There may not beany real objection to this, but the sugar obtained from it is like-wise termed a P-sugar, which is probably not the case.So much synthetical work on the sugars is based on the use oftetra-acetyl bromoglucose that i t is a matter of some importancetn have full details of a good working method for the preparationof this valuable reagent. For many years Fischer has prepared thecompound as the result of two separate reactions; first, the forma-tion of penta-acetyl glucose, followed by the conversion of thisP o J. C. Irvine and J. L. A.Macdonald, T., 1915,107,1701 ; A., i, 17.6 0 E. Anderson, J. Physical Chem., 1916,20, 269 ; A., i, 465.51 G. Pierce, J. Biol. Chem., 1915,23, 327 ; A., i, 1886 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.compound into tlie bromo-derivative by the action of hydrogenbromide dissolved in glacial acetic acid. He now gives a completeaccount of the method,52 but it may be remarked t h a t the processwhich has proved so conspicuously successful in the hands ofHudson and his collaborators possesses many advantages. Insteadof two stages, one is sufficient, in t h a t the reagent used-a solutionof hydrobromic acid in acetic anhydride-effects simultaneousacetylation and bromination. Many examples might be quoted toillustrate the convenience and efficiency of the method which, asrecently described,53 applies wit.11 equal facility to pentoses.54Considerable interest is attached t o the partial acylation of poly-hydric alcohols and sugars, which is now engaging the attentionof Fischer.55~ 56 These compounds have been prepared by thegeneral methods described by Irvine and Scott in the case ofreducing sugars, and by Irvine and Paterson for polyhydric alco-hols, and as few new examples have recently been added to thelist of partially substituted glucoses, most of the results are thusdescribed elsewhere.Although the generalisations established by Hudson regardingthe relationship between rotatory power and constitution in thesugar group were first applied to simple glucosides, it has recentlybeen shown in numerous cases that the principle extends to acetyl-ated aldoses of different types, and even to the monobasic acidsallied to tlie sugars.57 A further important observation is t h a t therule holds approximately in the case of acetylated derivatives ofamino-sugars, thus indicating t h a t the nature of the groups in thesugar chain has a relatively small effect on the rotation of theterminal asymmetric system.5* One result of this line of work hasbeen the preparation of a large number of acetylated sugars, andthe introduction of niaiiy improvements in the methods of prepar-ing and purifying these compounds.59 The results obtained in theoptical study of these compounds confirm Hudson’s views, and incases where experimental values are not in agreement with tlierule, it is probable t h a t this is due to the derivatives under esam-ination being related t o entirely different forms of the parent62 E.Fischer, Ber., 1916,49, 584 ; A., i, 373.53 J. K. Dale, J. Amer. Chem. Soc., 1916,38, 2187.54 J. K. Dale, ibid., 1915,37, 2745 ; A., i, 117.Part 11, E. Fischer and C. Rund, Zoc. cit.56 Part 111, E. Fischer and M. Bergmann, Zoc. cit.57 P. A. Levene, J. BioZ. Chem., 1915, 23, 145 ; A . , ii, 3.Levene and G. M. Meyer, ibid., 26, 355 ; A., ii, 545.5t3 C. S. Hudson and J. K. Dale, J . Amer. Chem. Xoc., 1916, 38, 1431 ; A.,i, 597.5 9 See also C. S. Hudson and D. H. Brauns, ibid., 1915, 37, 2736; &4.,i, 118 ; C. S. Hudson and J. M. Johnson, ibid., 1915,37, 2748 ; A., i, 117.See also P.AORG AN I C CH E M 1 STllP . 87sugar. Incidentally, in the course of this work a number of impor-t a n t observatioiis have been made. Thus, starting from tetra-acetyl fructose, f o r which a simple method of preparation isdescribed, a new forin of fructose penta-acetate is obtained by theaction of acetic anhydride containing zinc chloride. It is + curiousfact t h a t the two isonieric fructose penta-acetates so far known arenot intercoiivertible by the action of zinc chloride, and it is quitepossible t h a t they are derived from different types of fructose. Thechemistry of fructlose will no doubt be considerably expanded bythe use of tetra-acetyl fructose, and the reagent has already beenturned to good account in the preparation of a crystalline varietyof methylfructoside.60 There is little doubt t h a t fructose shows apronounced tendency to exist in reactive forms similar t o‘‘ y-glucose,” in which the ring-forming oxygen is connected t o acarbon atom other than t h a t in the y-position.Thus i t has beenshown61 t h a t the syrup originally isolated by Fischer by the con-densation of fructose with methyl alcohol contains two distinctmethylfructosides of different type. One of these compoundsresembles the a- or fl-methylglucosides, and fails to react withacetone, but the other enters into ready condensation with tneketone, and is, in addition, cliaracterised by the ease with whichit reduces permanganate. The existence of these distinct fruc-tosides clearly indicates that the parent sugar can react in a t leastfour forms, the most probable structures of which are shown below :OH.CH,*CH.[CH* OH];C(O H).C H,* OH ( Q and p)‘L-@--/Fructose of iiormd type.0 H . CH 2*[ CH.0 HI,* CH* C( OH) C 11 2* OH (a and p)\/0Fructose of ths ethylene oxide type,Additional evidence pointing to the idea t h a t fructose shows aready tendency to react in forms other than the standard a- and/3-varieties is aflorded by the somewhat complicated results recentlyobtained in a study of the conductivity changes shown during themutarotation of fructose dissolved in boric acid.62Brief reference was made in last year’s Report t o a new andconvenient method devised by Weerman for degrading sugars t olower members.Fuller details of the working methods employed,6o C. S. Hudson and D. H. Brauns, J . Amer. Chem. ,Sot., 1916,38, 1216 ;61 J. C. Irvine and G. Robertson, T., 1916,109, 1305.6a J. Boeseken, A. H. Kerstjens, and C . E. Klamer, Proc. K . Akad. Wetensch.A , , i, 547.A?nsterdam., 1916, 18, 1654 ; A . , i, 59688 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.and their range of application, are now available, and a few char-acteristic cases in which the method has been used successfully arenoted below.1. ~ - G ~ L I C O S ~ + d-arabinose.2. d-Galactose + d-lyxose.3. Z-Mannose --f Z-arabinose.4. Z-Arabinose + Z-erythrose.I n the experience of the writer of this Report, the reactionsinvolved 63 proceed with the utmost smoothness, and constitute agreat improvement on the methods hitherto in use.Among isolated observations which are of interest, the extensionto pentoses of the method of resolving racemic sugars through theagency of optically active mercaptals may be noted,6* and in viewof the number of successful applications of this process it may nowbe regarded as general in its application.Turning to an entirelydifferent subject, an observation which may in time prove to behighly important is t h a t the oxidation of glucose by means ofalkaline potassium permanganate is apparently assisted by atmo-spheric oxygen. This result is not altogether unexpected, but doesnot seem likely t o lead to any improvement in the methods avail-able for oxidising sugars t o t h s corresponding monobasic acids,although it may afford a clue to the mechanism of natural processesin which the sugars are converted into their ultimate oxidationproducts .65Although amino-sugars are few in number, they never lose ininterest.The complex problems connected with the constitutionof these compounds arO now being studied by Levene, who isengaged on a systematic investigation which has as its object thedetermination of the configuration possessed by glucosamine. Thework has not proceeded f a r enough for definite conclusions to bedrawn, but has resulted in a considerable extension of our know-ledge of the amic acids related to the amino-aldoses.666.67 Incident-ally, this work has confirmed a point of fundamental importancein this section of sugar chemistry, in that the suggestion alreadymade t h a t compounds of the type of glucosimine are in realityamino-glucosides, has been confirmed, and has been shown to beequally applicable to imines of the pentose series.68 Among other63 R.A. Weerman, “ Over de inwerking van natrium hypochloriet op amidenvan onverzadigde zuren en oxyzuren,” Dissertation, Delft, 191 6 .64 E. VotoEek and V. Vesely, Zeitsch. ZuclceTind. Bcjhm., 1916, 40, 207 ; A.,i, 308.65 E. J. Witzemann, J. Amer. Chem. Soc., 1916,38, 150 ; A . , i, 372.66 P. A. Levene, J. Biot. Chem., 1916,24,55 ; A., i, 203.67 P. A. Levene and F. B. La Forge, ibid., 1915,22, 331 ; A . , 1915, i, 944.6 8 P. A. Levene, ibid., 1916,24, 59 ; A . , i, 201ORGANIC CHEMISTRY. 89results recently to hand on this subject may be noted the fact thatchondrosamine and a lyxohexosamine prepared synthetically yieldthe same osazone, and, further, that the corresponding amino-acidsgive the same anhydrornucic acid on oxidation.69 The difficultieswhich usually attend the allocation of the amino-group in theseccmpounds may in some measure be removed by taking advantageof generalisations regarding the rotations of epimeric hexosamicacids, and the further extension of Levene's work will be closelyf0llowed.7~It is a remarkable fact that doubt still exists as t o the origin ofamino-sugars such as glucosamine, and as to the condition in whichthese compounds exist in the complexes from which they ar0 ob-tained by hydrolysis. The recent isolation of natural glucosamineglucosides from certain fungi seemed to indicate t h a t compoundsof similar type would be widely distributed and readily identified,but a careful examination71 of Boletus edulis has given resultsshowing t h a t the glucosamine formed on hydrolysis owes its originto complexes of the glucoprotein type, and not t o preformedglucosamine glucosides.Further work in this field will, therefore,doubtless be focussed on the isolation of glucoproteins of definitecomposition.As indicated in last year's Report, work on the synthesis ofglucosides by means of the action of enzymes has recently showna distinct falling off, but the new literature on this subjectis still fairly extensive. Among recent results of importancemay bO mentioned the synthesis of a galactobiose72 by the auto-condensation of galactose under the influence of emulsin.Con-sidering the fact that we are still far from a complete understand-ing of the linkings present in even thO simplest disaccharides, it isunfortunate that the new sugar, which is apparently different fromthe galactobiose already described, has only so far been obtainedas a syrup, and thus its detailed examination has naturally beenrestricted .Although exact studies on the selective hydrolysis of eitherglucosides or compound sugars ought to be conducted with enzymeswhich have been purified from closely related forms, it wouldappear that drastic purification is unnecessary in the case ofsyntheses effected by enzymes.73 It must be admitted that thesynthetical action of enzymes has in the past few years led tomany notable achievements, particularIy in cases to which ordinary6 9 P.A. Levene, J . Biol. Chem., 1916,26, 143 ; A . , i, 712.7 0 I b i d . , 367 ; A , , ii, 546.'l Miss W. Ross, Biochem. J., 1916,9, 313 ; A , , 1915, i, 1084.52 E. Bourquelot and A. Aubry, Compt. rend., 1916,163,60 ; A . , i, 596.'j3 l b i d . , 1915, 161, 463 ; A., 1915, i, 107690 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.processes are inapplicable, but despite this success the synthesis ofsucrose still remains unaccomplished,7* and, considering recent viewsas to the structure of sucrose, this negative result is not sur-prising.The analytical applications of enzyme action might perhaps bedealt with more appropriately in another section of the Report, butattention should be directed generally to the steady improvement,both as regards accuracy and range of application, of methodswhereby, through t h e use of selected enzymes, specific sugars canbe estimated even when they are mixed with closely related sub-stances.A good account of such a method is furnished by Davis,75who has shown, among other cases quoted, t h a t raffinose may beestimated by enzymokic hydrolysis. By a curious coincidence,practically the identical method was described about the same timefrom another source.76Disa cc hn rides.The hydrolytic process referred t o above, in which raffinose isdegraded to inelibiose and fructose, has been utilised for the pre-paration of t h e disaccharide on a large scale.The working detailsshow t h a t the process proceeds smoothly, and thus anotherexample77 has been added to the already lengthy list of excellentpreparations in the' sugar group, which we owe to the Americanchemists.From the theoretical point of view, two important contributionshave been made during the year to the chemistry of disaccharides.The first refers to the identification of numerical relationshipsbetween the molecular rotation of a disaccharide of the sucrosegroup and those of the a- and &forms of the aldose produced onhydrolysis. The interesting, but somewhat intricate, generalisationthus established can only be properly appreciated by reference t othe original paper,78 and, no doubt, its future development will beawaited with much interest.As a result of observations described in a recent paper,ig currentviews on the constitution of sucrose are largely overthrown.Thefact t h a t invert-sugar consists of an equimolecular mixture ofgluccse and fructose, has been too readily accepted as evidence t h a t74 W. Lob, Biochem. Zeitsch.. 1916,72, 392 ; L4., i, 296.75 J . SOC. Chem. Ind., 1915,35, 201 ; A., ii, 202.76 C. S. Hudson and T. S. Harding, J . Amer. Chon. Scot., 1915, 37, 2193 ;77 Ibid., 3734 ; A . , i, 120.7 8 C. S. Hudson, ibid., 1916, 38, 1566 ; A . , i, 630.7 9 W. N. Haworth and J. Law, T., 1916,109, 1314.A . , 1915, ii, 803.See also C. S. Hudsonand R. Seyre, ibid., 1867 ; A., i, 711ORGANIC CHEMISTRY.91the forms of the sugars thus isolated are of necessity the forms inwhich they are combined in the disaccharide. On the contrary,direct hydrolysis of sucrose affords little o r no evidence as t o theconstitution of the constituent hexoses, but a rigid proof has nowbeen supplied by a study of the hydrolysis of fully methylatedsucrose. Octaniethyl sucrose has now been obtained in the purestate in coiisiderable quantity, and gives, when treated with diluteacid, tetramethyl glucose and tetramethyl fructose, in which thelinkings are of necessity preserved exactly as in the parent disac-charide. The methylated aldose actually isolated proved to be thecharacteristic tetramethyl glucose of the butylene-oxide type,thereby confirming a previous result, but, on the other hand, theketose displayed a rotatory power and reaction towards perman-ganate which a t once stamped the compound as being allied instructure to " y-glucose." A new formula f o r sucrose is thus indi-cated, which accounts not only f o r the extreme ease with whichthe disaccharide is hydrolysed, but also for previous failure toeffect its synthesis.This provisional formula shows a butylene-oxidic aldose coupled t o an ethylene-oxidic ketose through theirreducing groups :CH,*OH/--O--\ \0H~CH,~CH(0H)*CH~[CH.OH],*CH~0~C-CH*[CH~0H]~~CH2*0H.\/0Accepting this view, the hydrolysis of sucrose becomes an exceed-ingly complex change, best represented by a diagrammatic scheme,which shows how the sugars liberated on hydrolysis not onlyundergo mutarot'ation, but, in the case of fructose a t least, changefrom one type t o another:Sucrose + a-glucose + a-fructose a-fructosep-glu cose p-fructose p-f ructose - v -( A ) .Butylene ( B ) Ethylene- (C) Butylene-oxide forms. oxide forms. oxide forms.The final products consist essentially of the equilibrium mixtures( A ) and (C) together, probably, with a small proportion of (13).The importance of the advance which has been made in tliestudy of the most interesting of all sugars, and the new lines ofwork thus opened out, will be fully realised92 BNNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Pol ysa ccha rid es.It is probably only right that much of the work in this groupremains physical rather than chemical in character, for there ismuch still to be learned as to the effect of surface conditions inmodifying chemical changes when dealing with complexes of highmolecular weight and sparing solubility. This is borne out inmuch of the recent work on the chemistry of starch, where, to takea case in point, the formation of the so-called crystalline starch80is in all probability due to the precipitation of the colloid by theaction of traces of mineral saltms.81 The influence of surface con-ditions is probably also responsible for the distinct' differencewhich has been noticed in the relative ease with which celluloseand starch may be acetylated, and in the effects which differentcatalysts have on the reaction.82The examination of starch by purely chemical methods has alsoyielded interesting results, and at' least one element of confusionhas been removed in the course of two recent investigations 839 84dealing with the phosphorus-containing constituent. The ideaseems t o have prevailed that the starch comple'x contains anunstable fragment, in which phosphoric acid is combined, and thatthis is the first portion of the molecule to be removed on hydro-lysis.Such is not the case, as during the degradation of starchwith 10 per cent. hydrochloric acid the unhydrolysed residue con-tains most of the combined phosphorus. In fact, there are goodreasons for the belief that a definite compound will in time beisolated which would represent the carbohydrate phosphate origin-ally present in the starch complex.The results of the paper nowunder review are not inconsistent with Samec's observation thatthe removal of the phosphorus constituent is the initial actionwhen starch is decomposed with alkalis.Naturally enough, most recent investigatims on cellulose havehad a direct technical bearing, and there does not appear t o beany necessity to refer to them, as no points of general theoreticalimportance have been noted.8 o M. W. Beyerinck, Proc. K. Akad. TYetensch. A,msterdam, 1915, 18, 305 ;A . , 1915, i, 940.W. Harrison, J. SOC. Dyers, 1916,32,40 ; A . , i, 251.82 J. Boeseken, J. C. van den Berg and A. H. Kerstjens, Rec. trav. chim.,1916,35, 320 ; A., i, 308.83 M. Samec, Koll. Chem. Beihefte, 1916,8, 33 ; A., i, 308.84 J. H. Northrop and J.M. Nelson, J. Amer. Chem. Xoc., 1916, 38, 472;A., i, 373ORGANIC CHEMISTRY. 93Nitrogen Compods.Most of the experimental researches of the past year which canbe suitably reviewed under this heading deal with topics whichare being expanded on normal lines, and there is little that canbe said regarding such publications except t o indicate isolatedobservations which may lead to useful improvements in workingmethods. Purely theoretical subjects have not been neglected, andreference has already been made to papers which deal with thevalencies of nitrogen, particularly with regard to the formationof conjugated systems when tervalent nitrogen is linked toethylenic carbon. I n addition, several other theoretical papershave appeared in which explanations are offered as t o themechanism of reactions conducted on nitrogen compounds,although it is unlikely that such views will find immediate orwidely spread application.85 The possibility of gleaning freshinformation as to the distribution of nitrogen valencies throughthe study of optically active compounds still continues to receiveattention, but efforts to isolate substances containing two similarlyasymmetric nitrogen atoms in active forms analogous t o thetartaric acids have led to no positive result. Even a modificationof this idea, in which the attempt has been made to resolve asystem containing two nitrogen atoms of unlike asymmetry, haslikewise proved fruitless, as, although a compound of the desiredtype has been obtained, no resolution was effected.86 The testsubstance selected for these experiments wasC A \ Ph-NI*CH,=CH2*CH2-NBr /Bz Me ,Me/ ‘Phand failure in this case, where the nitrogen atoms are separatedby a chain of carbon atoms, lends support to the belief that thesolution of the original problem wili not be realised consideringthe difficulty (or, it may be, the impossibility) of forming a com-pound containing two directly linked quinquevalent nitrogenatoms.87The double salt formulated above was isolated in two stereo-isomeric forms, and further examples illustrating the inexhaustiblecapacity of nitrogen compounds to form such isomerides isfurnished by a study of substances which contain two >C:Ngroups.88 Starting from the esters of carbazinic acid, alkylidene85 G.Povarnin, J . Russ. Phys. Chem. Soc., 1915,47, 989; A., 1915, ii, 761.86 E. Wedekind and T. Goost, Ber., 1916,49,942 ; A., i, 671.87 B. I(. Singh, T., 1916,109,780 ; A., i, 757.a8 M. Busch, J . pr. Chem., 1916, [ii], 93, 25 ; A., i, 33894 ANNUAL REPORTS ON THE PROGRESS O F CHEMISTRY.derivatives were obtained by condensation with aldehydes, and theproducts were thereafter decomposed with alcohols.NH2*NE-CS2R’ + CHR:N*NH-CS,R’ +CHR N N: C (S R’) SR ” .I n one case studied, all four possible stereoisomerides were isolated,and it may be noted that the last stage of the reaction affordsample scope fqr applying variations of Robinson’s theory of con-jugated systems involving tervalent nitrogen.Reactions of Simple Nitrogeib Coitzpoullds.A considerable amount of work has been devoted to developingimprovements in practical methods, and in some cases where thesedo not include any theoretical advances, it seems desirable t orefer to them, if only briefly. Thus, the method of alkylatingamines or imino-compounds by the agency of aldehydes in thepresence of an oxidisable compound has been extended consider-ably in scope and otherwise improved.89 To take another case ofimproved working processes, the observation 90 that aci-trinitro-methane, in the form of the potassium salt, is easily produced bythe interaction of tetranitromethane and potassium ferrocyanidewill exclude the older method of preparation, which, even inexperienced hands, proved to be extremely dangerous.After a period of successful use as a reagent, diazomethane hasrecently been but little employed, although it may be mentionedthat its action on aromatic acyl chlorides has been shown t o bevery similar to that on aldehydes, and gives rise to chloromethylchlorides.91R-COC1+ CH,N, -+- R*CO-CR,Cl.A t the same time, the application of diazomethane as a methyl-ating agent is evidently more restricted in scope than was a t onetime imagined.Hydroxyl groups of all types, and even carboxylgroups, are occasionally unaffected by the reagent, so that muchresearch of a fundamental nature remains to be done before itsuse in structural work will be fully assured. I n the case oftyrosine,92 the failure of diazomethane to methylate the carboxylgroup is probably due t o the existence of a betaine ring in themolecule, and it may be remarked, incidentally, that’ there is agrowing tendency to make use of cyclic structures containingquinquevalent nitrogen to explain reactions of simple compounds.Even in the case of guanidine, the suggestion has been put for-8 9 D.R.-P., 287802 ; A., i, 326.D.R.-P., 291222 ; A., i, 554. U.S. Pat.,1158496 ; A., i, 326.F. D. Chattaway and J. M. Harrison, T., 1916,109, 171 ; A., i, 245.91 D. A. Clibbens andM. Nierenstein, ibid., 1915,107, 1491 ; A . , 1915, i, 1062.92 A. Geake and M. Nierenstein, Biochem. J., 1915,9, 309 ; A., 1915, i, 1060ORGANIC CHEMISTRY. 95ward that the reactions of the compound are best expressed interms of such a cyclic structure,93 and although this is not opposedto many reactions of the compound, including its failure to givea positive result in the ninhydrin reaction,94 it apparently doesnot affect the deductions drawn by Levene and Senior as to theposition occupied by the two amino-groups in divicine.95A niiizo-acids and Proteins.Although research on amino-acids is once more reviving, thereis still little tendency to work with active products, and thenumerous pap?rs dealing with this subject, although contributingmuch valuable information, have not been fundamental incharacter, and are thus somewhat difficult t o discuss.So far assynthetical methods are concerned, little of importance has beennoted, but mention may be made of the formation of glycine frommethyl cyanoacetate by a series of reactions involving the produc-tion of cyanoacetylazide, CN*CH2*CO*N,, decomposition of thiswith alcohol, and hydrolysis of the product.As a method of pre-paring glycine, the reaction is of no importance, but is interestingas, taken in conjunction with the usual synthesis of the amino-acid from chloroacetic acid, i t constitutes an indirect proof of theinterequivalence of the four carbon valencies.96A considerable amount of work has been expended in studyingthe effect of nitric acid on piperazine derivatives.97 It is shown,as a result of these investigations, that nitric acid acts as anoxidising agent on 3 : 5-diketo-l-methylpiperazine,so as to generate the corresponding tetraketone, and that this typeof reaction seems to be general save when substituents are intro-duced into the two methylene groups.Thus, the imide obtainedfrom dimethyl iminodipropionate was not oxidised, but underwentnitration when treated with nitric acid, as shown below:>NH. CHMe-COCHMe-CO "<C CHMe*co>NH H Me* c m -+ NO,-N<It is unfortunate that the above imide was obtained only in smallamount, as the further effect of nitric acid on the compound is asubject which might well be expected to yield interesting results.$3 H. Krall, T., 1915,107,1396 ; A., 1915, i, 946.O4 V. J. Harding and R. M. MacLean, J. Biol. Chem., 1916, 25, 337 ; A.,96 P. A. Levene and J. K. Senior, ibid., 607 ; A., i, 67896 A. Darapsky and D. Hillers, J. p ~ . Chem., 1916, [ii], 92, 297 ; A., i, 127.9 7 J.V. Dubsky and pupils, Bey., 1916,49, 1037, 1041, 1045 ; A., i, 635,672,ii, 459.63606 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Despite the uncertainties which attend reactions carried outeither on amino-acids or on glycerol at comparatively high tempera-tures, work involving these possibilities continues to meet withsuccess. I n addition to the positive results previously obtainedin attempts to synthesise cyclo-glycylglycines by the somewhatdrastic method of heating an amino-acid with glycerol, it is satis-factory to note that mixtures of amino-acids react with equalfacility to give mixed cyclo-anhydrides, of which cyclo-alanylglycinemay be taken as a typical example.98 Greater interest will, how-ever, be taken in a recent communication from Abderhalden’slaboratory, which describes an ambitious synthesis,99 the ultimateobject of which is the production of polypeptides of high mole-cular weight so as to study the action of specific ferments andto determine the simplest molecular structure displaying colloidalproperties.Standard methods were employed in the synthesis,and as the essential reagent used was d-bromoisohexoyldiglycyl-glycyl chloride, the coupling reactions rapidly gave a polypeptidecontaining nearly a hundred atoms, the molecular weight of whichis more than 1200. It is surprising that the optical activitysurvived the reactions. As the uncertainties attending racemisa-tim or Walden inversion effects have in recent years greatlyrestricted synthetical work on active amino-acids and polypeptides,this revival will be welcomed, particularly as the work has adefinite objective.With regard to proteins, there is little to note in this section.The application of a drastic cleavage method involving the useof nitric acid does not seem to be advantageous save in specialcases, as, for example, the identification of phenylalanine throughits conversion into the easily recognisable pnitrobenzoic acid.1Among other results which have been obtained by the treatmentof proteins with powerful reagents, the most interesting is theobservation that, by the action of bleaching powder, proteins areconverted into derivatives containing the NC1 group, the presenceof which confers germicidal properties on the product.It wouldalso appear that both proteins and amino-acids react with chloro-amines? aldehydes being generally formed under these conditions,and work of this nature may possibly afford a clue to the anti-septic action of chloroamines or may in time be applied in struc-tural work on the proteins.3JAMES COLQUHOUN IRVINE.s8 L. C . Maillard, Ann. Chim., 1915, [ix], 4, 225 ; A . , i, 505.S S E. Abderhalden and A. Fodor, Ber., 1916,49,561 ; A., i, 375.C . T . Morner, Zeitsch. physiol. Chem., 1915,95, 263 ; A., i, 512.a H. D. Dakin, J. B. Cohen, M. Dufresne, and J. Kenyon, Proc. Roy. SOC.,1916, [B], 89,232 ; A., i, 533ORGANIC CHEMISTRY. 97PART II.--HOMOCYCLIC DIVISION.Reactions.DURING the past year o r two several novel reactions have beendevised.I n one, amines are oxidised to aldehydes by hydrogenperoxide in the presence of iron, typical examples being the con-version of ethylamine into acetaldehyde and of benzylamine intobenzaldehyde.1 I n another, aromatic aldehydes are prepared bythe action of carbon monoxide on the parent hydrocarbon, under apressure of 50-90 atmospheres, in the presence of aluminiumchloride ; benzaldehyde, ptolualdehyde, and p-chlorobenzaldehydehave been made in this manner.2 The introduction of theamino-group into aromatic compounds has been effected by con-densaticn with hydroxylamine in presence of sulphusic acid.3Sulphonic acid or nitro-groups of aromatic hydrocarbons can bereplaced by chlorine, in many cases quantitatively, by the actionof thionyl chloride a t moderately high temperatures; thus a t160-1 80° pchlorobenzenesulphonic acid yields pdichlorobenzeneand nitrobenzene yields chlorobenzene.4a-Arylamino-nitriles may be conveniently prepared by leaving asolution of an aldehyde o r ketone and the amine in glacial aceticacid with concentrated aqueous potassium cyanide ; thus, f o rexample, o-nitrobenzaldehyde and aniline yield o-nitro-a-anilino-phenylacetonitrile, N0,*C,H4*CH(NHPh)-CN.~ By the action ofdiazomethane on acyl chIorides chloromethyl ketones are ob-tained 6 :R*COCl+ CH,N, + R*CO*CH2Cl.Nitration.-In continuation of the work on the relative velocityof substitution caused by the halogen elements, chlorine, bromine,and iodine,' comparison of the directive influences of fluorine andchlorine has now been made, and it is found that in the nitrationof pfluorochlorobenzene the nitro-group enters the ortho-positionwith respect to the chlorine atom to the extent of 62-25 per cent.K.Suto, Biochem. Zeitsch., 1915, 71, 169; A,, 1915, i, 941.J. Longman, Eng. Pat., 1915, No. 3152; A., i, 316.J. F. d6 Turski, D.R.-P., 287756; A., i, 313.H. Meyer, Monntsh., 1915, 36, 719, 723; A., i, 134, 135; compare alsoKinzlberger & Co., D.R.-P., 280739; A., 1915, i, 658.i , 659.i. 1062.ii R. von Walther and R. Hiibner, J . pr. Chem., 1916,6 D. A. Clibbens and M. Nierenstein, T., 1915, 107,Ann. Reports, 1915, 89.REP.-VOL XIII.[ii], 93, 119 ; A.,1491; A., 1915,98 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.( ? 72.25 per cent.), and into the ortho-position t o the fluorine atomt o the extent of 27.75 per cent.*ReductioiL.-It is well known that the reduction of nitrobenzenet o aniline is carried out technically by means of iron and only1/40th of the amount of hydrochloric acid required by the follow-ing equation :Ph*NO, + 6HC1+ 3Fe = Ph-NH, + 3FeC1, + 2H,O.Many theories have been advanced to explain why the reactionproceeds after the iron has consumed the small amount of hydro-chloric acid present, and various sets of equations have been formu-lated t o represent the reduction as the result of a cycle of opera-tions. P.N. Raikow9 has recently reviewed these explanationsand examined a fundamental equation in each experiment’ally. I nRoscoe and Schorlemmer’s view,lo the initial reaction is followedby the reduction of nitrobenzene by means of ferrous chloride andhydrochloric acid :Ph*NO, + GFeCl, + 6HC1= PhoNH, + 3Fe,Cl, + 2H,O,but reduction in this way cannot be realised experimentally.0. N.Witt,ll who worked with a-nitronaphthalene, representedthe second stage of the cycle as the reduction of the nitro-com-pound by means of aqueous ferrous chloride:R*NO, + 6FeC1, + H,O = R-NH, + 3Fe,C1,0.This representation was criticised by A. Wohl12 on two grounds,namely, (1) that, since nitrobenzene can be reduced by means ofiron and aqueous calcium chloride, ferrous chloride is not essential,and (2) that nitrobenzene is not reduced by aqueous ferrouschloride a t the concentration employed technically.Raikow alsoconfirms the last result.Wohl put forward another explanation which has been acceptedby many writers of text-books. It is supposed that in the presenceof ferrous chloride moist iron reduces nitrobenzene direct with theformation of ferric hydroxide :Ph*NO, + 2Fe + 4H,O =Ph-NH, + 2Fe(OH),,and the latter then combines with ferric chloride to form a complexbasic salt. Objection has been taken t o this explanation on theground thatl the halogen is thus gradually withdrawn from solu-tion, when, in consequence, the vigour of the reaction shoulddiminish, whereas this is not the case. To avoid this objection,F. Swarts, Rec. trav. chim., 1915, 35, 131 ; A., i, 133.Zeitsch. angew. Chem., 1916, 29, i, 196, 239 ; A ., i, 469, 599.Ding. IpoZyyt. J . , 1887, 265, 225; A., 1857, 1048.Ber., 1894, 27, 1432, 1815; A . , 1894, i, 409, 450.lo “ Treatise on Chemistry,” 1891, Vol. 111, Pt. 3, p. 223ORGANIC CHEMISTRY. 99V. Meyer and, later, B6hal suggested modifications of Wold’sexplanation, in which purely “ catalytic” action is ascribed to theferrous chloride.A more satisfactory explanation is that given in Muspratt’sb00k,l3 namely, that the aniline and ferrous chloride formed in thefirst stage of the reaction interact to give ferrous hydroxide andaniline hydrochloride, the latter then reacting with metallic iront o liberate hydrogen or effect reduction:2Ph-NH2 + FeC1, + 2H,O = 2Ph-NH2,HC1 + Fe(OH)?,2Ph*NH2,HC1 + Fe= 2Ph*NH, + FeC1, + H,.Both these reactions can be confirmed experimentally, but a morefundament’al explanation of the process is put forward by Raikow.He attributes the continuation of the reduction primarily to hydro-lytic dissociation of the ferrous chloride.The hydrochloric acidso formed then reacts with metallic iron to liberate hydrogen oreffect reduction of nitrobenzene, regenerating ferrous chloride. Thereaction may then be represented as follows:Ph-NO, + 3Fe+ 6HC1= Ph=NH, + 3FeC1, + 2H,O.3FeC1, + 6H,O = [3Fe(OH), + 6HCll.[3Fe(OH)z -+ 6HC1]+ 3Fe + Ph*NO, =This explanation closely resembles the on0 put forward toexplain the reduction of nitrobenzene by iron and small quantitiesof acetic acid.14Catalytic Reduction.-When aromatic alcohols, aldehydes, andketones are reduced catalytically in acetic acid solution in ths pres-ence of colloidal platinum, they tend to yield hydrocarbons.Thusbenzaldehyde yields toluene and methylcyclohexane ; and benzo-phenone gives diphenylmethane and dicyclohexylmethane. If,however, the alcoholic, aldehydic, or ketonic grouping is suitably pro-tected, the hexahydro-derivative of the oxygenated substance maybe obtained. F o r example, if benzylideneaniline is first reduced t obenzylaniline by sodium and alcohol, and then submitted t o theabove catalytic process, dodecahydrobenzylaniline is obtained ; andthis, by oxidation to dodecahydrobenzylideneaniline and subsequenthydrolysis, yields hexahydrobenzaldehyde. Similarly, althoughphenylacetaldehyde yields P-cyclohexylethyl alcohol, together withsome ethylbenzene and ethylcyclohexane, on catalytic reduction, abetter yield of the alcohol may be obtained by reducing the acetylderivative of the aldehyde, phenylvinyl acetate, Ph*CH:CH*OAc.I n the case of cinnamaldehyde, the reduction leads first to y-phenyl-3Fe(OH), + 3FeC1, + 2H,O + Ph-NH,.l3 “ Technisch-chemisches Handbuch,” 1888, I, 942.14 N.V. Sidgwick, “ The Organic Chemistry of Nitrogen,” 1910, p. 46.E 100 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.propaldehyde, thoii t o y-pheiiylpropyl alcohol, and eventually toy-cyclohexylpropyl alcohol, with small quantities of propylbenzeneand propylcycZohexane.15Experiments have been carried out on the catalytic hydrogena-tion of organic compounds with non-noble metals a t normaltemperature and pressure.Good results have been obtained inaqueous-alcoholic solution, using, as catalyst, nickel precipitatedas basic carbonate on t o infusorial earth or other supporting sub-stmce, and reduced a t 450O.16Catalytic reduction has been applied to the estimation of halogenin organic compounds, from which in many cases the halogen isquai titatively eliminated by treatment with hydrogen in the pres-eiica of palladinised calcium carbonate. A long list of substanceswhich behave in this way is given; instances are, carbon tetra-chloride, iodoform, tricliloroacetic acid, chlorobenzene, pdibromo-ber,zene, and m-bromobenzoic acid.17OX~ddio72.-The study of the oxidation of phenols, t o whichreference was made last year,lB has been continued.Phenol itselfyields, on electrochemical oxidation, quinol, benzoquinone, catechol,4 : 4’-diphenol, 2 : 4’-diphenol, and o-hydroxydiphenyl ether. It isprobable that the diphenols are intermediate products in theformation of benzoquinone and dihydroxybenzenes, since they yieldthese substances on electrochemical oxidation.19 Derivatives of1 : 2 : 3 : 4-tetrahydroxybenzene have been prepared by the oxida-tion of derivatives of pyrogallol containing a free hydroxyl group;for instance, 2-hydroxy-3 : 4-dimethoxyacetophenone yields 2 : 5-di-hydroxy-3 : 4-dimethoxyacetoplienone on treatment with persul-phates in alkaline solution.20 I n view of the fact, previouslyknown, that salicylaldehyde yields gentisaldehyde (2 : 5-dihydroxy-benzaldeliyde) on oxidation with pessulphates, a knowledge of thebehaviour of its alkyl ethers and closely related compounds is ofinterest..It has now been found that anisaldehyde yields anisicacid when oxidised in this way, whilst. vanillin yields the diphenylderivative dehydrodivanillin. The latter may be obtained in excel-lent yield by the use of sodium prsulphate in the presence of alittle ferrous sulpliate.21By the oxidation of gallic acid with arsenic acid in sulphuric acid15 A. Skita, Rer., 1915, 48, 1685; A., i, 41.16 C. Kelber, ibid., 1916, 49, 55 ; L4., ii, 309.17 M. Busch and H. Stove, ibid., 1916, 49, 1063; A., ii, 534.18 Ann. Repor&, 1915, 94.Is F. Fichter and E. Brunner, Bull. SOC. chim., 1916, [iv], 19, 281; A.,2 0 G.Bargellini, Gazzetta, 1916, 46, i, 249 ; A., i, 489.21 K. Elbs and H. Lerch, J . pr, Chem., 1916, [ii], 93, 1 ; A., i, 315.i, 644ORGANIC CHEMISTRY. 101solution, a new colouriiig matter, flavogallol, has been obtained.The investigation of this interest,ing substance has not yet beencompleted, but the following formula is tentatively proposed 22 :OH 0 OHSteric and other Influences of Substitueftt Radicles.The influe,nce of different ortho-substituents on the readinesswith which diphenylmethane derivatives are oxidised to the corre-sponding benzhydrols by means of lead peroxide has been studied.?3I n 4 : 4/-tet7ramethyldiamino-Z : 2'-dimethyldiphenylmethane,the methyl groups only perinit the formation of the hydro1 to avery small extent, but their replacement by chlorine greatly dimin-ishes the steric hindrance.The influence of one o r more dimethyl-amino-groups in the ortho-position has now been examined, and itis found that the two following compounds undergo oxidation tothe corresponding hydrols :Me c1whilst. the two formulated below do not:Me Me MeAnother investigation deals with the cleavage of benzhydrols. Itwas shown some years ago that paminobenzhydrol and its deriv-atives react with bromine in a manner exemplified as follows p4 :OH*CHPh*C,H2Br2*NH, -+ Br, = Ph*CHO + C,H,Br,*NH, + HBr.It is now found that not only bromine, but also other reagents,such as nitric acid and nitrous acid, capable of effecting substitu-tion in the benzene nucleus, bring about a similar cleavage. Anumber of benzhydrols with a para-substituent have been examined,A.G. Perkin, T., 1916, 109, 529; A , , i, 485.23 J. von Braun, Ber., 1916. 49, 691 ; A . , i, 473.L. Clarke and (4. J. Esselen, jun., .T. Awler. C'hem. ,Voc., 1911, 33, 1135 ;101-1,36, 308 ; d., 10L1, i, 72.7 ; 1911, i, B i S 102 ANNUAL REPORTS ON THE PROGRESS O F CHEMISTRY.and it is found t h a t the readiness with which they undergo cleavagecorresponds with the efficiency of the para-substituent in promotingsubstitution in the benzene nucleus, the substituents examinedbeing effective in the following order:Me2N > OH > OMe > Me > Br.Further substituents in the ortho-position to these groups invari-ably diminish tlie relative amount of cleavage, and, with all but themost effective, diortlio-substitution stops it altogether.25The readiness with which ortho-substituted derivatives ofdimethylaminobenzene undergo different reactions has been com-pared.26 The reactions studied were (1) the introduction of form-aldehyde into the para-position, which yields dimethylamiiiobenzylalcohols, (2) tlie power of forming malachite-green-like dyes on con-densation with henzaldehyde and zinc chloride and subsequentoxidation, (3) the addition of methyl iodide, and (4) the oxidationof the diniethylaminobenzyl alcohols by means of formaldehyde t othe corresponding benzoic acids. The four reactions were studiedwith o-methyl-, o-chloro-, o-bromo-, and o-methoxy-dimethylamino-benzene under identical conditions, and the following comparativeresults were obtained :0 rt ho -subst ituen t Me.c1. Br. OMo.Reaction 1 ............ 6 36 45 GO per cent.2 ............ Negative. Negative. Trace. Positive.4 ............ 5 1 G 36............ cn36 per cant.3 7-6 15.6 16%It was found, further, that the 1 : 8- and 1 : 6-dimethyl-1 : 2 : 3 : 4-tetrahydroquinolines (I and 11) were much more reactive than0- and p-dimethylaminotoluenesCH,/\I 1 1 I N M ~ ,\./\/cH2 \ /Me NMe Me( I a and I I a ) respectively:CH,An interesting study of the relative quantities of ortho- andpara-nitrophenyl esters obtained by tlie nitration of diff ereiit phenylesters has been carried out. The aiiiounts of the o-nitrophenylester were as follows : with triphenyl phosphate, 5.6 ; diphenyl phos-phate, 16.0 ; pheuyl phosphate, 16.3 ; phenyl carbonate, 10.7 ; andphenyl methyl carbonate, 14-9 per cent.In the case of plienylstearate, iione of the o-nitroplienyl ester appeared to be formed,26 E. P. Kohler and R. H. Patch, J . .4uicr. Chr?n. Sot., 1916, 38, 1205 ;21i J. von Braun, Bcr., 1916, 49, 1101 ; d., i, G-17A . , i , 557ORGANIC CHEMISTRY. 103whilst phenyl acetate gave 20-25 per cent. of the o-nitrophenylester cor,taminated with some of the 2 : 4-dinitrophenyl e~ter.~7The following is an example of the combined influence of certaingroups in preventing the6-nitro-m-toluonitrile (I) 28Me')NO,CNI \/reaction of another group. Neithernor 5-nitro-o-toluonitrile (11) 29 can beiMeON/>\,NO2(11.)reduced to the corresponding amino-nitrile. This is apparently duet o the combined influence of the, para-cyano-group and the alkylgroup on the stability of the nitro-group, f o r the carboxylic acidson the one hand, and pnitrobenzonitrile on the other, can bereadily reduced. Incidentally, it may be of interest to note thatthe oxidation of 5-nitro-o-toluonitrile, which may be prepared from5-nitro-o-toluidine by the Sandmeyer reaction, is a ready methodfor the preparation of 5-nitrophthalic acid.Alkyl Ethers of Polyhydric Phenols.A very large number of naturally occurring substances are deriv-atives of alkylated polyhydric phenols. Most of the alkaloids con-taining a quinoline or isoquinoline nucleus are substituted in thebenzene portion of the molecule by hydroxy-, methoxy-, ormethylenedioxy-groups.A knowledge of the properties of thecorresponding benzene derivatives which are likely to be formedin the degradation of such substances is therefore of great import-ance in the determination of constitution. In his comprehensiveinvestigation of cryptopine, W. H. Perkin, jun.,3O has obtained alarge number of degradation products of this type, many of whichwere previously unknown. I n some cases the determination of theconstitution of the degradation product has been a matter of con-siderable difficulty, as, for instance, in the case of the compoundCgH,O,. This had the odour of piperonal, and gave an oxime,semicarbazone, and phenylhydrazone, and, therefore, appeared t obe a methyl derivative of piperonal, having the formulaCH,:O,: C,H2Me- CHO.On oxidation, i t yielded a methylpiperonylic acid,CH,:O,: C,H,Me - CO,H,27 (Mlle.) J.M. A. Hocflake, Rec. trav. d i m . , 1916, 36, 24 ; A . , i , 472.? * Bcilstein and Rreusler, Annalen,, 1807, 144, 175.2 p M. Mayer, J. pr. Chem., 1915, [iiJ, 92, 137 ; A . , 1915, i, 958.3 0 T., 1916, 109, 816104 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.but this could not be oxidised to the dicarboxylic acid, which wouldbe either identical (I) or isomeric (11) with hydrastic acid:CO,H(1.1 (11.)The methylpiperoiiylic acid was therefore converted into the cor-responding dimethoxymethylbenzoic acid by first forming di-hydroxymethylbenzoic acid under the conditions recommended byG.Barger 31 for the hydrolysis of a niethylenedioxy-group :HO/\ \and subsequefnt methylation. This acid gave a good yield of hemi-pinic acid (111) on oxidation, and therefore had one) of the twoformulz (IV) or (V):Me C0,H C0,HMe0 (\Mef- MeO\/ \/(111. ) (L'. )MeO/\CO,Zl --+ MeO/\CO,HMe01 MeO1 1 \/(IV. 1It was next found t h a t the dihydroxymethylbeiizoic acid wasdecomposed by heating with water a t 170-180° with eliminationof carbon dioxide and formation of a dihydroxytoluene. Thisproved to be isoliomocatechol (VI), and not homocatecliol (VII),and consequently the hydroxymethylbenzoic acid from which itwas derived had the formula (VIII), and the methyl derivative ofMe Me aleHOI IO/\CHOHO/\ HO/)lle HO/\CO,H CW,<Ol IHOI Hob \/(TS.)\/(VIII.) (VII.)\/(VI.1piperonal, of which the constitution was sought, had the forinulaOther new derivatives of polyalkyloxybeiizeiies, described in thisvery interesting paper, include 4 : 5-dimethoxy-2-vinylbenzene (X),which is closely related t o hydrastal (XI), and m-opianic acid(XII), both of which were obtained from cryptopine, ando-hydroxy-3 : 4-dimethoxy-o-tolylacetalclehyde (XIII) with the cor-responding acid (XIV), which were obtained from berberine .(IX).M~O/\CHO O/\CHO M~O/\CHOM e d IC0,HMeO!,,!CH:CH, c H 2 < ~ \ , b ~ : ~ ~ , \/(ST.)31 T., I!)OS, 93, 57ORGANIC CHEMISTRY. 105/\CH,* CHO /)CHz*C02H\/(XIII.) (XIV.)Me01 ,!CH2*OH M~O!,/CH,-OHMe0 Me0A knowledge of the nitro-derivatives of 2 : 3-dimethoxytolueneand 2 : 3-dimethoxybenzoic acid is of importance in connexion withtwo naturally occurring substances, the dyestuff santalin, andhydrourushiol (1 -n-pentadecyl-2 : 3-dihydro~ybenzene)~ a constitu-ent of Japanese lac.2 : 3-Dimethoxytoluene yields on nitration first5-izitro-2 : 3-dimethoxytoluene, which gives 5-nitro-2 : S-dimethoxy-benzoic acid on oxidation, and then 5 : 6-dinitro-2 : 3-diinethoxy-tolueiie :Me Me Me3-Hydroxy-2-methoxytoluene, however, yields first a mixture ofthe 4- and 6-mononitro-derivatives, and then 4 : 6-dinitro-3-hydroxy-2-methoxytoluene~ :Me Me Me MeBy the oxidation of the inethyl ether of the 4-nitso-derivative thehitherto unknown 4-nitro-2 : 3-dimethoxybenzoic acid has been ob-tained. The proofs of the constitutions of the above substancesare of considerable interest, and the original papers32 should beconsu lted .It may be noted that whilst syringic acid (I) in cold acetic acidsolution yields 4 : 5-dinitropyrogallol 1 : 3-dimethyl ether (11) onnitration, its methyl ester yields in acetic anhydride at - - 5 Omethyl 2-iiitrosyringate (III).33OH OH OHMe0f)OMe MeO/\OMeMeO{)OMe No"\/ NOZj, iC02Me\/(111.)C0,H NO2(1.1 (11.)32 R.Mrtjima and Y. Okazaki, Ber., 1916, 49, 1482,and Sci. Rep. TdhokuI m p . Univ., 1916, 5, 215 ; A . , i , 808 ; compare also J. C. Cain and J. L.Simonsen, T. 1914, 105, 156.m M. T. Bogert and E. Plant, J . ,4mer. Cliem. SOC., 1915, 37, 2723; A.,i , 146.P106 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.F.Reverdin's 3.1 observation that piodoanisole yields on nitra-tion o-iodo-p-nitroanisole, with apparent migration of the iodineatom, has led Mrs. G. M. Robinson35 t o study the mechanism ofthis reaction. She expresses the view that the reaction is not duet o migration of the iodine atom, but depends on the liberation ofiodine and its subsequent entry int,o the molecule in the ortho-position. o-Iodo-p-nitroanisole is doubtless formed to some extentas follows:HO OMeHOI + H I = I, + H20OM0 OMebut the process mainly responsible for its formation is probablythe following :OMe OMe OLMesince di-iodoanisole can actually be isolated as an intermediateproduct of the reaction, and can be converted into o-iodo-p-nitro-anisole by further treatment with nitric acidDynamic Isomerism.The discovery that the two isomeric hydrojuglones are keto-enol isomerides36 has led to the investigation of the reduction pro-duct of naphthazarin (I), which easily gives a tetra-acetyl deriv-ative, and is designated 1 : 4 : 5 : 6-tetrahydroxynaphthalene (11).CO OH CO/\/\GH /\A /\f\CH2HO GO HO OHHO!,/,,CH~HO COHO(,),,!'CH(1.1 (11.1 (111.)s4 Ber., 1896, 29.1003 : A., 1S96, i, 475.a5 T., 1916, 109, 1078; A., i, 805. 36 Ann. Reports, 1915, 104ORGANIC CHEMISTBI’. 107It is now found37 that this substance readily yields a phenyl-semicarbazone aud a phenylliydrazone, and there is some evidenceof the formation of a diphenylsemicarbazone, but hydroxylamineand semicarbazide decompose it, owing to their alkalinity, so t h a tderivatives of these reagents could not be prepared.1 : 4 : 5 : 6 -Tetrahydroxynaphthalene is therefore regarded as reacting accord-ing to the two formuh I1 and 111, and thus resembles phloro-glucinol, which apparently exists only in one form, but exhibitsboth ketonic and phenolic properties.Chromoisomeris M.Hypotheses designed t o explain the existence of a substance informs of more than one colour have been plentiful during recentyears. Hantzsch, as is well known, depicts the differently colouredvarieties as structural isomerides, the difference in structuredepending either on the arrangement of the normal linkings oron the arrangement of the subsidiary valencies. I n the case ofhelianthin and the aminoazobenzenes, an account of his viewswas given in a previous Report,39 and these have since beenelaborated by their author in further papers.39Hantzsch40 has now observed chromoisomerism in the case ofcompounds free from nitrogen, namely, in substances of the ethylacetoacetate type.Symmetrical compounds of the general formulaR*CO=CH,-CO*R, for example, acetylacetone and indandiones,give monochromic salts only, but asymmetric compounds of thegeneral formula R*CO.CH,*CO*R’ give in many cases chromo-isomeric salts, which may be formulated R-C(OM):C€I*CO*R’ andRmCO*CH:C(O&I)R’ respectively (where M represents the metal) ;or, when the residual affinity between the metal and the un-saturated carbon atom is taken into account, they are given theconjugated formulaeInstances of the formation of chromoisomeric salts have beenobserved with ethyl succinylsuccinate, ethyl dihydroxybenzo-quinonedicarboxylate, and ethyl 2 : 5-dihydroxyterephthalate, andin each case the chromoisomerism is attributed to the existence37 A.S . Wheeler and V. C. Edwards, J . Amer. Chem. SOC., 1916,38, 387;38 Ann. Reports, 1913, 107.39 Ber., 1915, 48, 158, 167; A., 1915, i, 321, 322.40 Ibid., 785, 797 ; A., 1915, i, 549, 551.A . , i, 392.E’ 708 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.of the salt in two structurally isomeric forms, which may berepresented as-fi*OM and -?I0 ON-C*CO,R -C:C<ORBenzenoid or phenolic salt. Quinonoid or enolic salt.I n the case of ethyl dihydroxyterephthalate, not only do thesalts exist in two chromoisomeric forms, but, also the free ester,which occurs in a colourless and a yellow variety.The colour-less form is optically similar to ethyl dimethoxyterephthalate, andmust therefore be the true dihydroxy-ester, whilst the yellow formis assumed to have the quinonoid or enolic formula.The same author puts forward a similar explanation of thechromoisomerism of certain salts of phenolic aldehydes, namely,that they are to be formulated as follows:MO- C,H4*CH0 and 0 :C,H4: CH OM,the paler-coloured salt having the phenolic and the darlrer-colouredsalt the quinonoid-enolic formula. This suggestion has led to acontroversy41 with H. Pauly, who contests many of Hantzsch’sstatements and refers t o the colour of the salts of substitutedm-hydroxybenzaldehydes. These have only been observed in oneform, but some colourless m-hydroxybenzaldehydes yield colourlesssalts, whilst others yield yellow salts.The existence of theseyellow salts is held t o refute Hantzsch’s view that the deep colourof the salts of phenolic aldehydes is due to quinonoid structure,for there is an increasing volume of evidence against the existenceof meta-quinonoids. Pauly contests Hantzsch’s view that thechromoisomerism of phenolic aldehydes is due to desmotropy, pre-ferring to regard the phenomenon in the light of the valence-electron theory.That the occurrence of chromoisomerides may often be attributedsimply to polymorphism has been pointed out previously42 inthese Reports, and this view is now supported by P.Pfeiffer,43 whooffers an ingenious speculation on the internal structure of certainchromoisomerides. He finds that, nitromethoxystilbenes containingthe methoxy-group in the para-position, and the nitro-group in thesecond benzene nucleus, as, for instance, 4-nitro-4/-methoxy-stilbene, N0,.C6H4*CH:CH=C6H,*OMe, occur in two forms, ayellow and an orange, one of which is usually stable at lower andthe other a t higher temperatures. Both forms have the, same41 A. Hantszch. Ber., 815, 1332; 1916, 49, 234; A., 1915, i, 551, 1062;1916, i, 403; H. Pauly, ibid., 1915, 48, 934, 2010; A . , 1915, i, 689; 1916,i, 150.42 Ann. Reports, 1911, 54-60; 1913, 252.43 Ber.. 1915, 48, 1 7 7 7 ; A., i, 24ORGANIC CHEMISTRY.109melting point and give solutions identical in colour. The nitro-methoxystilbenes give greenish-yellow solutions in benzene, butorange solutions in trichloroacetic acid, and, where crystallineadditive products have been isolated, those with benzene and aceticacid are yellow, whilst those with substances having a higherdegree of residual affinity, such as trichloroacetic acid and stannicchloride, are orange. The removal of the second component ofsuch additive compounds is found t o cause a change of colour;thus, the addit,ive compound of 2-nitro-4-benzoylnmino-4~-methoxy-stilbene and acetic acid is yellow, and on gentle heating loses aceticacid, leaving the orange variety of the stilbene, whilst, the additivecompound of the same stilbene and trichloroacetic acid is orange,and, on removal of the acid by the same means, yields the yellowvariety of the stilbene.I n order to explain these results, Pfeiffer has recourse to histheory of halochromy,44 and to the knowledge recently gained ofcrystal-structure by the researches of von Laue and the Braggs.The phenomenon of halochromy in the case of ketones depends onthe co-ordinative addition of acids or metallic salts to thecarbonyl-oxygen atoms of the ketone ; this addition causes analteration of energy in the ketone molecule, so that the carbonyl-carbon atom becomes more or less unsaturated, and thereforechromophoric in character.R,C:O + HX -+ R,C:O...HX.The coloured additive products of nitro-compounds with phenolsor phenolic ethers, R-NO,.. . C,H,*OH, probably owe their colourt o a similar cause. It is now suggested that the different colouredforms of the nitromethoxystilbenes are caused by the differentarrangement of the single molecules in the crystals. Where thenitro-group of each stilbene molecule is combined co-ordinativelywith the residual affinity of an unsaturated carbon atom of anothermolecule, coloured compounds, similar to the additive compoundsof nitro-compounds and phenolic ethers, will be produced. Suchcompounds will have the darker, orange form. In the yellow formit is supposed that the arrangement of the single molecules is dueto polymerisation similar to that occurring with nitroso-compounds.Very roughly, the idea may be represented as follows:??0,*C&&gCH:CHgC6H4 *OMe N0,*C,H4.CH:CH*C,H,*OMeMeO*C,H,*CH :CR-C,H,-NO, NO,* C,H,*CH:CH* C,H,- OMeOrange.Yellow.The curious colour reversal noted in the case of 2-nitro-4-benzoyl-amino-4/-methoxystilbene is then explained as follows. The orange-44 Anizakn, 1911, 383, 92 ; A , , 1911, i, 788.This is expressed as follows :110 ANNUAL REPORTS O N THE PROGRESS OF CHEMISTRY.coloured solution of the trichloroacetic acid compound containsthe co-ordination compound R*NO,. . .HO*CO-CCl,, in which theresidual affinity of the oxygen atom of the nitro-group is alreadysaturated by the trichloroacetic acid molecule. There is, there-fore, no tendency for this substance to crystallise in the form inwhich the nitro-group is co-ordiiiatively combined with an uii-saturated carbon atom of another molecule of the stilbene.It issupposed to crystallise, therefore, according to the scheme offormation of the yellow crystals of the free stilbene, and the factt h a t the substance is actually orange in colour is due to the halo-chromic combination of the trichloroacetic acid molecule ; on theremoval of this, the yellow form of the free stilbene remains. Thestructure of the yellow acetic acid compound is such t h a t thestilbene molecules which it contains are arranged as in the orange-coloured stilbene crystals, b u t the acetic acid molecules saturatemore or less completely the free colour-causing affinities and giverise to the paler (yellow) form; on the removal of the acetic acid,the orange form of the free stilbene crystals appears.It is perhaps of interest in connexion with Pfeiffer’s work tonote the existence of 2 : 2’-dinitro-4 : 5 : 4’ : 5’-tetramethoxyazo-benzene,NO, NO,Med-\N :N/-\OMe, \ / \--/MeO- OM0in two varieties, forming red and yellow needles respectively. I nthis case, however, the separate colours are preserved after the twosubstances have been dissolved in sulphuric acid, recovered bydilution with water, and recrystallised, and it is thought t h a t theanomalous colour of one of them is due to a small proportion ofa probably isomorphous impurity.45Another interesting suggestion as to the cause of chromo-isomerism in a special case is p u t forward by I.Lifschutz andF.W. Jenner.46 pNitrophenylacetonitrile forms two isomericsodium salts, one of which gives a red alcoholic solution, and isthe more stable t o alkalis, whilst the other gives a green alcoholicsolutdon, and is obtained by leaving the red solution for some timeor treating it with carbon dioxide. The spectra of the red andgreen salts are identical in the ultra-violet and analogous in thevisible region, and therefore the green salt, like the red,47 must46 (Mrs.) G. M. Robinson and R. R.obinson, T., 191.5. 107, 1753 ; &4., i , 166.46 Ber., 1915, 4$, 1730 ; A . , i, 45.47 Compare J. T. Hewitt, F. G. Pope and (Miss) .W. I. Willett, T., 1912,101, 1770ORGANIC CHEMISTRY. 111have a quinonoid configuration. The authors indicate a differencebetween the two salts by representdng them as stereoisomerides,thus :H CNO:N:/=\:d O:"/--\:dI \-/ I I * \ = = = / IOM CN OM Hpreferring this explanation to one involving the residual affinitiesof the nitro- and cyano-groups in a 7- or &membered ring.Triar y lin e t h y 1.A number of interesting reactions of sodium triphenylmethylhave been described.This substance cannot be condensed withany compound capable of reacting in an enolic form, since thesodium atom is then replaced a t once by hydrogen, but, subjectt o this limitation, it can be condensed with esters to yield ketones,and aldehydes to yield alcohols. Thus, methyl benzoate yields/3-benzpinacolin,Ph,CNa + MeO*COPh = Ph,C-COPh + MeONa,formaldehyde gives triphenylethanol, CPh,*CH,*OH, and benz-aldehyde tetraphenylethand, CPh,*CHPh*OH.The reactions ofsodium triphenylmethyl are thus similar to those of Grignard'sreagents. Sodium triphenylmethyl combines with sulphur dioxideto yield sodium triphenylmethylsulphinate, CPh,*SO,Na, but doesnot react with carbon monoxide. With ammonia it yields tri-phenylmethane and sodamide.48 It reacts with tetramethyl-ammonium chloride to give triphenylmethyltetramethylammonium,an interesting compound in which the nitrogen atom is linked tofive hydrocarbon radicles.49CPh,Na + ClNMe, = CPh,*NMe, + NaCl.Polycyclic Aromatic Hydrocarbons.1ndene.-A convenient method for the preparation of indonesof a certain type has been described. It had previously beenshown t.hat the interaction of benzophenone, ethyl a-bromo-propionate, and zinc led to the formation of ethyl B-hydroxy-BB-diphenyl-a-methylpropionate,PhCO + CHMeBr*CO,Et + Zn = ZnBr*O*CPh,*CHMe*CO,Et,ZnBr*O*CPh,*CHMe*C02Et + H,O =ZnO + HBr +OH=CPh,*CHMe*CO,Et,which could be simultaneously hydrolysed and dehydrated toB-phenyl-a-methylcinnamic acid by treatment with diluted4 8 W.Schlenk and R. Ochs, Ber., 1916, 49, 6 0 8 ; &4., i, 379.4 9 W. Schlsnk and J. Holtz, ibid., 603; A . , i, 386112 ANNUAL REPORTS OM THE PROGRESS OF CHEMISTRY.sulphuric acid in boiling acetic acid.chloride, this acid gave 3-phenyl-2-methylind0ne,5~Under the action of thionyl/, ./ \--gPh OH*CPh,*CHMe*CO,Et + CPh,:CMe*CO,H J \/\PMe' coIt is now found that ethyl j?-hydroxy-Pj?-diphenyl-a-metbyl-propionate readily gives this indone in good yield when treatedwith.cold concentrated sulphuric acid,51 only the one operationbeing necessary. Similarly, ethyl P-hydroxy-OD-diphenyl-u-ethyl-propionate yields 3-phenyl-2-ethylind0ne.5~The method of synthesising indandiones by the condensation ofdiethylmalonyl chloride with aromatic compounds in the presenceof aluminium chloride has proved useful in confirming the con-stitution of certain benzenetetracarboxylic acids. The condensa-tion product with pxylene, namely, 4 : 7-dimethyl-2 : 2-diethyl-indan-1 : 3-dione (I), gives mellophanic acid (11) on oxidation, whilstthe condensation product with m-xylene gives prehnitic acid (111) .53Me CO,H(1.1 (IT. 1 (111. )These syntheses confirm the constitutions previously assigned t othe acids.54NaphthaZene.-In connexion with the constitution of purpuro-gallin and lapachol, which yield naphthalene on distillation withzinc dust instead of 4-methylnaphthalene, as might be expectedfrom the formuh assigned to them, the study of a synthetic com-pound of similar type is of interest..2-Hydroxy-l-keto-4-methylene-1 : 4-dihydronaphthalene (11), which was prepared by the action of0 0\/\/C(C0,Et ),(1.)50 H. Rupe, H. Steiger, and F. Fiedler, Ber., 1914, 47, 63 ; A., 1914, i, 281.51 R. de Fazi, Atti R. dccnd. L i n c e i , 1915, [v], 24, ii, 343 ; A., i, 151.52 R. de Fazi, ibid., 1915, [v], 34, ii, 150; Gazzetta, 1915, 45, ii, 143 ;63 M. Freund and K. Fleischer, Arznalen, 1916, 441, 14 ; A ., i, 317.54 A. Baeyer, ibid., 1873, 166, 325 ; A . , 1873, 755 ; H. Bamford and J. L.A , , 1915, i, 1063.Simonsen, T., 1910, 97, 1908ORGANIC CHEMISTRY. 113pyridine on 2-hydroxy-1-keto-4-dicarbethoxymethylene-1 : 4-dihydro-naphthalene (I), also yields naphthalene when distilled with zincdust, the methylene group being eliminated.55 This result servesto remove a possible objection t o the formulze proposed for thenatural compounds of, as yet, uncertain constitution.The three possible dihydro-P-naphthoic acids, having the carb-oxylic group in the reduced ring, have the following formulz:(A1.> (AZ.) (A3.)Two of these are obtained by the reduction of P-naphthoic acidwith sodium amalgam, namely, a labile acid which is capable ofresolution into optically active components, and is theref ore theA3 acid, and a stable acid, for which the A2 formula has been pro-posed.It has now been found that both acids can be convertedinto a third isomeride by means of aqueous barium hydroxide a t160-180°. The dibromides of the three acids have been preparedand decomposed by means of cold 10 per cent. aqueous sodiumcarbonate ; under these conditions, the new acid yields a dihydroxy-acid, the so-called stable acid yields 8-naphthoic acid, whilst thelabile acid yields a monobromolactone. The formulze of these pro-ducts are shown below:C H,--YH, CH:QHc6H4<CH(OH)*C(OH)*C0,H C6H4<CH:C*C0,HThe results leave no doubt as to t,he constitutions of the threedihydro-8-naphthoic acids. Confirmation of the A1 formula for thenew acid has been obtained by oxidising it t o o-carboxyphenyl-propionic acid.56Similar work on the two known dihydro-a-naphthoic acids hasbeen published.57S~zthmcene.-Some interesting work has been carried out inconnexion with the constitution of kermesic and carminic acids,the colouring principles of kermes dye and cochineal respectively.Previous work5s has shown that kermesic acid has one of thefollowing formula :55 H.F. Dean and M. Nierenstein, T., 1916, 109, 593 ; A., i, 555.66 C. G. Derick and 0. Kamm, J. Amer. Ch.em. SOC., 1916, 38, 400; A.,57 0. Ksmm and H. R. McClugage, ibid., 419 ; A . , i , 395.5 5 Ann. Reports, 1913, 132.i , 394114 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.0 0CO,H 11 OH Me 11 OHHO/\/'/\CO Me / \ /\/\(jOj&\/\/\/ I I I IOH H d \ANoH IMe 11 OH CO,HII O H0 0(1- 1 (TI.)and experiments are now recorded of which the double object wasa preliminary study of syntheses in this field and an attempt todecide which of the above formulae was correct.It is shown thatpurpurin (1 : 2 : 4-trihydroxyanthraquinone) may be synthesised bythe condensation of phthalic anhydride and hydroxyquinol (in theform of its triacetate). Similarly, 4-hydroxyphthalic anhydrideyields with hydroxyquinol a tetraliydroxyanthraquinone, whichmight be either the 1 : 2 : 4 : 6-derivative, hydroxyflavopurpurin (111),or the 1 : 2 : 4 : 7-derivative, hydroxyauthrapurpurin (IV) :0 OH+\/OH/\ x(111.)W.)The product proves t o be the former substance, for i t is identicalwith the oxidation product of flavopurpurin, 1 : 2 : 6-trihydroxy-anthraquinone.Since kermesic acid closely resembles hydroxy-anthrapurpurin (IV) (obtained by the oxidation of anthrapur-purin) in its absorption spectrum and tinctorial properties, whilsti t differs from hydroxyflavopurpurin in these respects, i t is con-cluded that formula I correctly represents the relative positionsof the hydroxgl groups in kermesic acid.59Carminic acid appears t o be closely related t o kermesic acid,but of more complex constitution. Amongst its properties, thecapacity of yielding, by degradation, derivatives of naphtha-quinone is peculiar, and experiments have now been made todetermine whether purpurin (V), which i t resembles in tinctorial5 * 0.Dimroth and 12. Fick, _4nnrrh, 1916, 411, 315 ; A., i, 561ORGANIC CHEMISTRY. 115properties, can be made to undergo a similar degradation. Thesehave given an affirmative result, for purpurin yields Z-hydroxy-3-acetyl-a-naphthaquinone (VI) when oxidised with alkalinehydrogen peroxide in the presence of cobalt oxide.60(V. 1Phenanthren.e.-An ortho-hydroxyaldehyde has been preparedOf the two possible from 3-phenanthrol by Gattermann's method.(VII.) (VIII.)formuh for this substance, VII and VIII, the first has now beenestablished by converting the substance into morphol (3 : 4-dihydr-oxyp henant hrene) .6lHydro cyclic Corn pounds and T erpe 12 es.By the action of zinc dust on the compound C(CH,Br), inaqueous-alcoholic solution, G.Gustavson G2 obtained a hydrocarbonwhich he regarded as vinyltrimethylene (I). Addition and sub-sequent elimination of hydrogen iodide transformed it into anisomeride, which was represented as ethylidenetrimethylene (11).CH,*CH: c<FH2C*2(1. ) (11.1These compounds have since been the subject of several in-vestigations, which left their constitut,ion doubtful, but0. Philipov 63 has now shown that they are methylenecyclobutaneand methylcyclobutene respectively.c a,: c<g:;>c H,6 o 0. Dimroth and E. Schultze, Annulen, 1916, 411, 339 ; A., i, 563.61 J. W. Smith, T., 1916, 109, 568; A., i, 487.c2 J . p r . Chem., 1996, [ii], a, 98, 105 ; with Miss H. Bulatoff, ibid., 1897,[ii], 56, 93 ; .4., 1896, i, 6 6 9 ; 1898, i, 13.69 Ibz'd., 1916, [ii], 93, 162; A., i, 551.Compare also N. J. Demjanov,RcT., 1908, 41, 915: A.. 1008, i, 329, and A. Favorski and W. Batalin,ibid., 1914, 47, 1648: J . Russ. Ph.ys. CFem. Soc., 1914, 46, 726; A . , 1914,i, 815; 1915, i, 390116 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.The evidence supporting this view is based on the results of theoxidation and reduction of these compounds. On oxidation,methylenecyclobut ane gives cyclobutanone, together with a mix-ture of acids, including glutaric acid, the unexpected formationof which he explains by the following scheme:CH2*Q0cH <CH2>C(oH)'CHo CB, --t CH2<CH2.CH,0H -+CH, C0,HOxidation of methylcyclobutene leads to laxulic and succiiiicacids, as follows:cH2<EE>CH -+ CH2<_CH2>CH*OH CMe(0H) +COMe*CH,*CH,*CO,H ---+ C2H,( C02H),.An identical reduction product was obt.ained from the twohydrocarbons, and this has been identified as methylcyclobutaneby comparison with a specimen of this substance preparedsynthetically by Kishner's method, as indicated below :CH,<Eg:>CH*CHO + CH2<:E2>CH-CH:N*NHz 2 +N.Kishner,64 in a further study of his method of forminghydrocyclic compounds, shows that 5-phenyl-3-styrylpyrazoline (I),obtained by the action of hydrazine on distyryl ketone, cannotbe converted into a derivative of cyclopropane, but, yields, on heat-ing, 3 : 4-diphenylcyclopentane (11).W HCH2-v HPh S1HPh.C H2CHPh-CH CHP h: CH- C<N-NH(1.1 (11.1The closely allied pyrazoliiie (111), however, when heated in thepresence of po,tassium hydroxide and platinised porous tile, losesnitrogen, forming l-phenyl-2-~-phenylethylcycZopropaiie (IV).CH2Ph*CH2*C<N-N CH,*YHPh --+ CH,Ph* C H,aCH<? H2CHPh(111.) (IV.1A method for the conversion of cyclohexanones into cyclo-pentanones has been elaborated, and many examples of its usehave been described.65 As an example, its application t o cyclo-64 J . Russ. Phys. Chem. Soc., 1915, 47, 1819; A . , i, '290.65 0. Wallach, M. Gerhardt, and W. Jessen, Nnchr. I<. Ges. IPiss. G6ttingen,1915, 244 ; A., i, 487ORGANIC CHEMISTRY. 1 1 7hexanone itself may be recorded. On bromination in acetic acidsolution, this ketone yields a dibromide, which yields a ketone,C,H,O,, and a hydroxycarboxylic acid, C,H,,O,, on treatmentwith dilute aqueous potassium hydroxide at the ordinary tempera-ture. The hydroxy-acid decomposes into cyclopentanone andcarbon dioxide when distilled with lead peroxide and sulphuricacid, and therefore has the formula IV.The ketone, C6H802, isrepresented by the formula 111, and is probably formed bydehydration of the Bypothetical intermediate product (11).(1.) (11- 1 (111.) (IV. )Camphene.-Now that the constitution of camphenic acid (I)been established by P. Lipp’s synthesis,”, probable formulae canbe assigned t o some of its derivatives. On dry distillation theacid yields camphenonic acid, a ketonic acid, which, since it canbe reconverted into camphenic acid by fusion with potassiumhydroxide or treatment with sodium and alcohol, is represented byformula 11.CH,*CH*CMe, CH,*CH-CMe,CH,-CH CO,HI I OH4 -A -7- I Y H 2 1CH,-C--COIC0,H &O,H(1.1 (11.)The d- and Z-forms of camphenonic acid have also been obtained,in each case together with the &variety, from the camphenicacids derived from camphenes of strong dextro- and lmo-rotatorypower respectively.67Camphene hydrate, which is readily obtainable by the action ofdilute alkali on camphene hydrochloride, can be reconverted, bygentle dehydration, into camphene having the original specificrotatory power.The hydrate is therefore formed without causinga change in the camphene skeleton. Of the two formulae, I and11, which might represent it, the second is elimina.ted by thefurther evidence that camphene hydrate and its hydrochlorideCH,.CH-CMe, CH,*CH-CMe, CH,. CH- CMe,CH2*CH-CMe-OH CH2-CH-CH*CR,-OH CH,*CH-CO(1.1 (11.) (111. )I I I 1 YH2 I I YH2 I I YH2 I66 Ann. Reports., 1914, 119.67 0. Aschan, dnnalen, 1915, 410, 2 4 0 ; A . , i, 52118 APUKUAL REPORTS OX THE PBOGHESS OF CHEMISTRY.show the typical behaviour of tertiary alcohol and chloriderespectively.Camphene hydrate, therefore, has the formula I, which alsorepresents methylcamphenilol, a substance obtained by the actionof magnesium methyl iodide on camphenilone ( I I I ) . G S The twocompounds, camphene hydrate and methylcamphenilol, are not,however, identical, but must be regarded as stereoisomeridesforming a pair comparable with borneol and isoborneol.Bothcompounds yield camphene on dehydration.69The previously known chloride of hydroxymethylenecamphormay be prepared readily by the action of thionyl chloride onhydroxymethylenecamphor. When the chloride is treated withGrignard's reagents, the chlorine atom is replaced by an alkyl orThe constitutiofi of the products obtained follows from the facts(1) t h a t the phenylmethylenecamphor so produced is identicalwith the previously known benzylidenecamphor resulting from theaction of benzaldehyde on sodium camphor, and (2) that- the pro-ducts yield with ozone some camphorquinone besides camphoricacid.70Amongst other investigations of hydroaromatic compounds,attention may be directed t o the following: an attempt to corre-late the constitution and the physical properties of isomeric andhomologous hydroaromatic compounds, ref erring mainly t o cyclo-hexane, cyclohexene, cyclohexanol, and cyclohexanone and theirmethyl derivatives ; 71 investigation of the constitutions of the foursaturated bicyclic hydrocarbons, C10H18, to which the namefenchane has been given;72 and a research on the synthesis ofterpineols and terpins.73Cholesterol.The earlier work on cholesterol was fully reviewed byA.Windaus74 in 1908, and his work on the subject has since beendealt with in the Annual Report for 1912.75 I n the meantime,68 S. Moycho and F. Zienkowski, Ber., 1905, 38, 2461 ; A . , 1005, i, 654.6 9 0. Aschan, Annalen, 1915, 410, 222 ; A , , i, 51.7 0 H.Rupe and M. Iselin, Ber., 1916, 49, 25 ; L4., i, 409.71 K. von Auwers, Annalen, 1915, 410, 287 ; -4., i, 130.72 S. S. Nametkin, J. Russ. Phys. Chenz. Xoc., 1915, 47, 1590 ; 4 . , i, 269.73 0. Wallach and collaborators, h'ach~. K . Ges. W'iss. GBttingen, 1915, 1 ;74 Arch. Pharm., 1908, 246, 117.A . , i, 213.75 Pp. 107, 109ORGANIC CHEMISTRY. 119steady progress has been made in the study of the constitution ofthis complex substance.The formula (I) favoured by Windaus a t the later date hasproved capable of explaining the more recently published develop-ments of the work, but, in the latest paper by one of Windaus’spupils,76 the modified formula (11) is put forward, for reasonswhich are t o be given later./\CH CH‘17’32 /\CH CH(1.1 (11.1The question as t o whether cholesterol contains one or twodouble linkings seems now to be answered.The stability of thetetrabasic acid, C21H3006, prepared from cholesterol step bymakes it improbable that the latter contains more than one doublelinking.78 Moreover, the molecular refractions of the substanceand of many of its derivatives are in close agreement with thosecalculated on the assumption that only one double linking ispresent in the molecule of chole~terol.~S The fact that it combineswith six atoms of oxygen when treated with washed ozone of lowconcentration indicates that two double linkings are concerned inthe reaction, but it seems probable, as C . Dor6egO suggests, thatonly one double linking exists preformed in the cholesterol mole-cule, and that the secondary unsaturation is due to the openingup of a bridged ring under the action of ozone.This view hasalso been adopted by 0. von Fiirth and G. Felsenreich.81During the last few years considerable progress has been madein determining the mutual relations of the reduction products ofcholesterol. It has been shown that the supposed dihydro-cholesterol, a-cholestanol, the product of the action of sodium andamyl alcohol, or of preformed sodium amyloxide, on cholesterol,is, in fact, an isoamyl derivative of cholesterol.82 P-Cholestanolis doubtless the normal dihydro-derivative of cholesterol, for it can7G T. Westphalen, Ber., 1915, 48, 1061; A . , 1915, i, 884.77 A. Windaus, Zbid., 1908, Pi, 2566 ; A ., 1908, i, 728.7 8 Compare C. DorBe, T., 1909, 95, 653.7 g L. A. Tschugaev and P. Koch, Annalen, 1911,385, 352 ; A., 1912, i, 30.LOC. cit. ; with L. Orange, T., 1916, 109, 46 ; A., i, 261.Biochem. Zeitsch., 1915, 69, 416; A., 1015, i, 679.A. Windaus and C. Uibrig, B e y . , 1913, 46, 2487 ; A., 1013, i, 969120 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.be prepared from the latter iii nearly quaiititative yield by reduc-tion with hydrogen in the presence of platin~m-black.~~8-Cholestanol suffers partial rearrangement when heated withsodium amyloxide and amyl alcohol, yielding 10 per cent. of astereoisomeride, termed E-cholestanol, which can also be convertedinto the equilibrium mixture in the same way. UnlikeP-cholestanol, echolestanol is not precipitated by digitonin, andthis property serves for the separation of the two isomerides.Theisomerism of these alcohols is caused by the steric arrangement ofthe hydroxyl group and hydrogen atom about the asymmetriccarbon atom t o which they are attached, for both alcohols yieldthe same ketone on oxidation.** A similar relation has beenshown t o exist between coprosterol and $-coprosterol.*s Animportant advance in our knowledge of the constitution of copro-sterol is due to the discovery86 that the saturated hydrocarbon,coprostane, derived from coprosterol, is identical with $-cholestane,which was prepared from cholesterol by J. Mauthner.87 Windausgives the following interpretation of the reactions carried out byMauthner.Replacement of the hydroxyl group of cholesterol(111) by hydrogen gives cholestene (IV), which may be reduced toP-cholestane (V). The addition product (VI) of cholestene andhydrogen chloride gives chiefly the isomeric hydrocarbon,I CH bH bHCH(VI. (VII.) (VIII. )83 R. Willstatter and E. W. Mayer, Ber., 1908, 41, 2199 ; A., 1908, i, 636.84 A. Windaua and C. Uibrig, ibid., 1914, 47, 2384; A., 1914, i, 1066.*6 C. Dorhe and J. A. Gardner, T., 1908, 93, 1630.86 A. Windaus and C. Uibrig, Ber., 1915, 48, 857 ; A . , 1915, i, 675.87 Monntsh., 1909, 30, G35 ; A . , 1909, i, 714ORGANIC CHEMISTRY. 121$-cholestene (VII), on the removal of hydrogen chloride, and$-cholestene yields on reduction mainly $-cholestane (VIII).I n the conversion of cholestene (IV) into JI-cholestene (VII),the asymmetric carbon atom to which the methyl group is attachedloses its asymmetry, and on reduction of $-cholestene to$-cholestane (VIII), the asymmetry of this carbon atom isregained, when the formation of an isomeride is readily under-stood.It follows that fl-cbolestanol and coprosterol, the alcoholscorresponding with the hydrocarbons fl-cholestane and $-cholestane(coprostane),. are similarly related. The relations of these twoalcohols with E-cholestanol and $-coprosterol may be representedas follows:B-Cholestanol.A\dE€ Mk!HCH, 6H2\/ CH,IICHAH6 - Choles tanol.I (?HCopros t erol.-1-i--AHJ/-Coprosterol.The preparation of a structural isonieride of cholesterol,+-cholesterol, may be noted,@ and also the formation ofa-cholestantriol by the conversion of the ethylen ic linking ofcholesterol into an ethylene-oxide group, and subsequenthydrolysis .a9-CH:CH- + -CH*CH- + -CH(OK)*CH(OH)-.\/0Cholesterylsulphuric acid, C,7H4,-O*S0,H, has also beende~cribed.9~88 A. Windaus and C.Resau, Be).., 1915,128, 851 ; A , , 1915, i, G77.8 9 T. Westphalen, ibid., 1064; A . , 1915, i, 884.J . A. hlnl-rdel anclC'. Senberg, Riochem. Zcifech., 1915, 71, 186 ; A . , 1916,i, 9 5 i 122 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Aromatic Compouiids of nitrogen.T h e Constitution of Acoxy-compounds.--It was pointed out ill190291 (I) that our knowledge of the azoxy-compounds was veryincomplete, and that the commonly accepted formula R-N-N*R\/0lacked experimental foundation, for the known reactions of thisclass of compound could be explained equally well by the formulaR*N:NO*R. Later work has brought to light properties of azoxy-compounds which can only be explained by the second formula,and has thus served t o establish it.A. Angeli’s discovery92 thatazobenzene gave azoxybenzene when oxidised with hydrogenperoxide in glacial acetic acid solution, was the first step leadingto this result, for this method when applied t o asymmetricallyconstituted azobenzenes gave rise t o two isomeric azoxybenzenes,which were not interconvertible, and differed not only in theirphysical, but in their chemical properties-f o r example, theirbehaviour towards bromine and nitric acid. The following ex-ample will serve to illustrate the work which has been carried o u ton many pairs of isomerides.93 Oxidation of p-bromoazobenzenewith hydrogen peroxide in glacial acetic acid solution yields amixture of a-p-bromoazoxybenzene, m.p. 7 3 O , and P-p-bromoazoxy-benzene, m. p. 9 2 O . When treated with cold bromine, theP-isomeride yields 4 : 4’-dibromoazoxybenzene, whilst the a-com-pound is unacted on under these conditions,94 but yields2 : 4-dibromoazoxybenzene under the influence of bromine and irona t 115-120O.95 This behaviour indicates the constitutionalformulae given below for the two isomerides, since it is probablethat substitution takes place in the nucleus united t o the tertiarynitrogen atom rather than in that united to the quinquevalentnitrogen atom.(a) CGH,* NO: N* C,H,Br + CGH,*NO : N* C,H,Br2.( p ) C,R,=N:NO*C,H,Br + C,H,Br*N:NO*C,R,Br.The oxidation of azo- to azoxy-compounds is thus analogous t othe oxidation of tertiary amines t o their oxides, R,N+ R,N:O.T h e Constitution of Hydrozymo-cornpounds.-Discussion of theconstitution of hydroxyazo-compounds has been renewed with91 V.Meyer and P. Jacobsen, “ Lehrb. d. organ. Cham.," 1902, 11, i, 251.g2 Ann. Reports, 1910, 98.9s Compare numerous papers by A. Angeli and B. T’alori, 1910-1915.94 A. Angeli and B. Valori, Atti R. Accad. Lincai, 1912, [vl, 21, i, 155.95 B. Valori, ibid., 1013, [v], 23, ii. 125 ; -1., 1913, i, 1110.A., 1912, i, 321ORGANIC CHEMISTRY. 123reference to a particular case.Anthraquinone does not form aphenylhydrazone by the direct’ method, but a compound may beobtained by the action of phenylhydrazine on 9 : 9-dibromo-anthrone :NHPhONH, + CBr2<$:4>CO + NHPh*N:C<:6:4>CO. 6 46 4The same substance results from the action of benzenediazoniumNINPhCl+ CH&C,H4>C*OH + NPh:NmC<2z>C-OH. (A)I<. H. Meyer and K. Zahn 97 regard the compound as benzeneazo-anthranol (A4), basing their view on the following comparison of itsproperties with those of benzeneazoanthranyl benzoate,chloride on anthranol96 :‘CoH‘I 6 4N P h N* C<(=GH4>C *O Bz (B) ,( 3 2 3 4and anthraquinonephenylbenzoylhydrazone,NPhBz-N:C<2g4>C0 (C) :6 4A . R. C.Colour ....................................red dark red yellowStability towards hydrolysis.. ....... great great smallReaction with bromine ............... immediate immediate slowFormation of halochromic additiveproducts ........................... positive positive negativeWhilst the O-benzoate was readily obtained by benzoylation ofthe parent substance, alkylation led to the formation of the 1Y-sub-stituted derivative, f o r example, anthraquinonephenylrnethylhydr-azone. Meyer and Zahn explain this difference in behaviour on theassumption that benzoylation is effected by a simple double de,com-position, R-ONa + ClBz + R*OBz, whilst alkylation is the resultof addition and subsequent fission, thus :NaO NaO I 0 \ ....... \/ II\/\/ \/\/ \/\/1 I +CH,I + II II + II II/\A /’\/ \I INN IIP h i .....NPhkMe Ph?kMeThey do not, however, leave out of consideration the possibilityt h a t the compouucl is tautomeric in Laar’s sense of the word, t1ia.tis to say that it is a single substance with two possible formulae, but5G F. ICaufler and W. Suchannek, Eel.., 1907, 40, 618 ; A . , 1907, i, 226.97 Annalen, 1913, 396, 152 ; t l . , 1913, i, 537124 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.they reject this view in favour of representing the substance asbenzeneazoant hranol.G. Charrier has now attacked the constitution of this substance,and arrives a t the opposite view, namely, that it is anthraquinone-phenylhydrazone. I n order to follow his evidence on the subject,it is necessary, first of all, t o refer briefly t o the operation whichhe terms " diazo-scission." Hydroxyazo-compounds (or theirethers) combine with two molecules of nitric acid t o form salts,which on heating undergo " diazo-scissioii,JJ that is, conversion intoa nitrophenol (or a nitrophenyl ether) and a diazonium nitrategg:NO,*NHAr:N* C,H,*OH,*NO, =Incidentally, it may be remarked that the dinitrates of aminoazo-compounds similarly yield, on fission, nitroarylamines and aryl-diazonium nitrates, a reaction which Charrier 99 represents as adecomposition of the dinitrate, whilst L.Casale 1 considers that themechanism consists first of dissociation of the nitrate into theaminoazo-compound and nitric acid, followed by reaction betweenthem. Charrier 2 finds that the diazo-scission characteristic ofhydroxyazo-compounds fails completely in the case of the so-calledbenzeneazoanthranol.Moreover, the conditions for the hydrolysisof anthraquinonearylhydrazones are very similar to those of anthra-quinonearylalkylhydrazones, which are known to have the hydr-azone formula.3 Taking into consideration the whole of the resultsdescribed by the above authors, it appears to be probable that thecompound in question really reacts in both the possible forms.The Comtitution of the Nitrosoliydrazon48.-When aldehyde-hydrazones are treated with nitrous acid or amyl nitrite, nitroso-derivatives are obtained, which can be rearranged by pyridine intoazoaldoximes (111). For the nitroso-derivative, two formulae arepossible, representing tl10 nitroso-group as attached t o nitrogen (I)and carbon (11) respectively :ArN:N*C,H,*OH + 2HN03 =NO,*NAriN + OH-C6H,*NO2 + H,O.RON( NO) *N :CH-R (I.)R-N:N*CP\:N*OH (111.)'x71R*NH-N:CR*NO (11.) '98 Compare, for instance, G.Charrier and G . Perreri, Gazzettn, 1914, 44,9 9 Ibid., 1914, M, ii, 503; A . , 1915, i, G G .1 Ibid., 1915, 46, ii, 397 ; A., i, 225.2 Atti R. Accad. Rci. Torino, 1914-1916, 50, 580; Gazzetta, 1915, 45,3 L. Omarini, i M . , 1015, 45, ii, 80-1 ; A . , j, S7. G. C'linrritlr, A t t i R. -4cc</d,i, 165 ; A,, 1914. i, 599.i, 502; A., 1915, i, 904.Sci. T o r i m , 181G, 51, 572 ; A., i. 511ORGANIC CHEMISTRY. 125E. Bamberger and W. Pemsel4 adopted the second formula forthe nitroso-derivatives on various grounds, such as analogy tof ormazyl compounds and the nature of the decomposition productsof the substances, but mainly because the nitroso-derivatives can beoxidised to nitroaldehydehydrazones, R-NH=N:CR*NO,, the nitro-group of which is undoubtedly united t o a carbon atom.More-ov0r, whilst they were unable to prepare a nitroso-derivative ofbenzaldehydephenylmethylhydrazone, they succeeded in preparinga nitro-derivative, PhMeN*N:CPh*NO,, of this compound. Theyconsidered and rejected as improbable the view that nitrosohydr-azones had the formula (I), and were able t o react in the dynami-cally isomeric form (11). M. Busch and H. Kunder5 have raisedthis question again, and conclude that these nitroso-derivatives aretrue nitrosoamines. They find that whilst benzaldehydephenyl-hydrazone yields a nitroso-derivative when treated with a nitrite inacetic acid solution, benzaldehydephenylmethylhydrazone does notdo so.Moreover, ketohydrazones also yield nitroso-compounds,which are in all essential particulars similar to the nitroso-aldehyde-hydrazones, and must be true nitroso-amines, R-N(NO)*N:CR2,since there is no alternative formula in this case. These factsindicate that the essential condition for the formation of a nitroso-derivative of this type is the presence of an imino-group, and thatthe presence of a hydrogen atom attached t o the aldehydic carbonatom is unnecessary.Complex Salts of Nit rosoarylhydroxyla mines.-Nitrosophenyl-hydroxylamine, PhN(N0) *OH, forms with iron, copper, titanium,and zirconium internally complex salts, to which the following con-stitution has been attributed 6 :,PhN PhN(where M is a univalent metal).These salts are characterised by abnormal colour and ready solu-bility in organic solvents.Metals other than those mentionedabove yield only normal salts of nitrosophenylhydroxylamine. 0.Baudisch7 and his collaborators have now commenced a study ofthe influence of substituents in the phenyl nucleus of this com-pound. Introduction of the group *C1, *Br, *NO, *Me, or -0Medoes not alter the selective action of the nitrosoarylhydroxylamine4 Ber., 1903, 36, 57, 359; A., 1903, i, 283, 286.Ibid., 1916, 49, 317 ; A., i, 436.8 E.Bamberger and 0. Baudisch, Ber., 1909, 42, 3577; A., 1909, i, 9777 Ibid., 1916, 49, 172, 180, 191, 203; A ., i, 386, 387, 388, 389126 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.in forming internally complex salts, for, in these cases, only thesalts with iron, copper, titanium, and zirconium are of internallycomplex nature. The circumstances are quite different when thesubstituent group is -O*SO,-C,H,Me, *OH, *hTH.SO,*C,H,b~e,*NMe-SO,-C,H,Me, *NMe,, or -CHO in the ortho- or meta-position,f o r here the capability of forming internally complex salts isextended to nearly all the metals. To give an example, o-nitroso-hydroxylaminophenyl 4-toluenesulphonate,OH=N(NO)*C,H,.O*SO,*C,H~Me,forms salts of this type with iron, copper, nickel, cobalt, chromium,manganese, vanadium, lead, cadmium, mercury, bismuth, alumin-ium, cerium, lanthanum, and thorium.Baudisch points out thatthe further developments of this work may have considerable im-portance for analytical and physiological chemistry.0zimes.-The isomerism of the oximes has been discussed a tlength,B and it is suggested t h a t the oximes are tautomeric sub-stances capable of reacting in the forms shown below:\ / JY Y Benzantialdoxime. Benzsynaldoxime.Certain properties of the oximes are more readily explained bythe assumption of an isomerism in this sense than by Beckmann’srepresentation of a tautomerism as between the oxime and isooximef ormuh :C,X,*CHH O=NI/ andBenzantialdoxime.I n t,he course of this work i t is shown that phenylcarbimide isnot a suitable reagent for the determination of the configurationof the oximes, since it yields with anti-oximes a certain amountof t h s carbanilino-syn-derivative, and there is some evidence t h a tthis is, indeed.the primary product of the reaction. On hydrolysis,the carbanilino-syn-derivatives of the oximes behave similarly t othe acyl derivatives, the sy n-derivatives yielding a nitrile and theanti-derivative the anti-oxime :CO, + PhNH, R-CH /-? I/ + R-CN + C0,H-NHPhN-O* C0.N HPh ‘* CO(NHPh),+NH,Ph+CO,.HO-N+ NHPh*CO-0-Ntakes place on similar lines.The formation and degradation of the carbethoxy-derivatives8 0. L. Brady and F. P. Dunn, T., 1916, 109, 650 ; L4., i, 651ORGANIC CHEMISTRY. 127MisceZZaneous.-A new and interesting method of formation ofazoimides is as follows : the aryldiazonium derivatives of trinitro-methane readily decompose in the presence of moist ether into aryl-azof ormhydroxamic acids, which yield the corresponding arylazo-imide, carbon dioxide, and hydrogen when heated with aqueousalkali hydroxide 9 :Ar*N,*O*ON:C(NO,), -+Phenylhydrazine is a useful reagent for effecting the reductionof azo- and bisazo-compounds to the corresponding amines, for theaction proceeds rapidly in the cold or with gentle heating, andgood yields of pure products may be obtained.Examples of its useare the reduction of benzezeazothymol to 6-aminothymol, bisazo-carvacrol to diaminocarvacrol, [OH : Me : Pr : (NH2)2 = 2 : 1 : 4 : 3 : 51,and benzeneazosalicylaldehyde to the phenylhydrazone of 5-amino-2-hydroxykf;nzaldehgde.loArN:N*CO*NH*OH + ArN3+ C02+ H2.Organometallic Compounds.Bntimony.-The interaction of triphenylstibine and antimonytrichloride in xylene a t 240° has been studied by many authors.J.Hasenbaumer 11 obtained phenylstibine dichloride in this way,and represented the reaction as follows:Ph3Sb + 2SbC1, = 3PhSbC1,whilst A. Michaelis and A. Giinther 12 obtained diphenylstibinechloride as the result of the reaction. G. T. Morgan and (Miss)F. M. G. Micklethwait13 found that both substances were pro-duced, and regarded the condensation as a balanced reaction inthe sense indicated by the following equation :(la) 2Ph3Sb + SbC13 = 3Ph2SbC1 ;( l b ) Ph,SbCl+ SbCl, 2PhSbC1,.G. Griittner and M. Wiernikl4 have studied the reaction again,with important results.They have shown that phenylstibinedichloride is gradually converted into diphenylstibine chloride andantimony trichloride when distilled under diminished pressure,2PhSbC1, -+ Ph,SbCl+ SbCl,,so that it is impossible to separate phenylstibine dichloride anddiphenylstibine chloride quantitatively by fractional distillation.I n order t o investigate the composition of the product at differentG. Ponzio, Gazzetta, 1915, 45, ii, 1 2 ; 1916, 4.6, ii, 5 6 ; A., 1915, i, 1011,1916, i, 609.lo E. Puxeddu, ibid., 1916, 46, i, 62, 211 ; A., i, 292, 435.l1 Ber., 1898, 31, 2910; A., 1899, i, 209.l2 Ibid., 1911, &, 2316; A., 1911, i, 1056.l3 T., 1911, 99, 2286. l4 Ber., 1915, 48, 1749; A., i, 96128 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.stages, it was necessary to devise a method for converting the con-stituents of the mixture into derivatives which would not undergochange.This was accomplished by treating the mixture withmagnesium ethyl bromide, when antimony trichloride gave tri-ethylstibine, phenylstibine dichloride phenyldiethylstibine, anddiphenylstibine chloride diphenylethylstibine, whilst triphenyl-stibine remained unchanged. The alkylstibines were thenseparated by fractionation distillation, the amount of each wasestimated, and from the results the composition of the originalmixture was calculated. The authors conclude that the reactiontakes place in two stages, the first proceeding most readily andquickly.(2n) Ph,Sb +- SbCl,= PhSbC1, + Ph2SbC1 ;( 2 b ) Ph2SbC1 + SbCl, = 2PhSbC1,.The fact that the use of a larger proportion of antimony chloridedid not increase the yield of phenylstibine dichloride affordsevidence against Morgan and Micklethwait’s suggestion of anequilibrium (1 6 ) .It may be noted that the reaction between triphenylbismuthine(2 mols.) and bismutl tribromide (1 mol.) in ether also gives riset o both diphenylbromobismuthine and phenyldibromobismuthine.~5Lead.-The study of mixed lead tetra-alkyls has been under-taken with the ultimate view of preparing asymmetric derivativesof lead.16 It is found that lead tetramethyl and lead tetraethylcan be converted into lead trialkyl haloids by treatment with asolution of the halogen in carbon tetrachloride at -75O, and thatlead trialkyl haloids react readily with magnesium alkyl haloids toyield mixed lea& tetra-alkyls :R,Pb + R,PbCl+R’MgCl + R,R’Pb.With one exception, all the lead trimethylalkyl and lead friethyl-alkyl compounds so prepared react with halogen under the aboveconditions and exchange one of the three methyl or ethyl groupspresent for halogen, thus yielding lead dimethyl-(or diethy1)-alkylhaloids, Me,RPbCl ; the exception is lead triethylmethyl, whichyields lead triethyl haloids.It is hoped that the lead dimethyl-(or diethy1)-alkyl haloids will react with Grignard’s reagents togive lead tetra-alkyls, Me,RR’Pb, containing three different alkylgroups. A mixed lead tetra-aryl, namely, lead diphenyldi-o-tolyl,PbPh2(C6H,Me),, has been obtained by the action of lead diphenyl-di-iodide on magnesium o-tolyl bromide.17l5 F.Challenger, T., 1916, 109, 250 ; A., i, 347.18 G. Griittner and E. Krause, Ber., 1916, 49, 1125; A., i, 684.17 K. Lederer, ibid., 349; A., i, 446ORGANIC CHEMISTRY. 129Sodium .-The preparation of disodio-deriva t ives of compoundscontaining a C:C, C:N, or K N linking, for examples, of stilbene,beiizylideiieaiiiliiie and azomethane, is recorded in the patent litera-ture.18 On treatment with water, the alkali metal is replaced byhydrogen, whilst’ carbon dioxide converts the compounds into thecorresponding, carboxylic acids.Tellurium.-A number of diaryl ditellurides and diaryl tellurideshave been prepared by the action of magnesium aryl haloids ontellurium dihaloids, and their properties, which resemble those oft,he selenium analogues, have been studied.19FRANK LEE PYMAN.PART III.-HETEROCYCLIC DIVISION.THE continuance of the war has again made the compilation ofthe Report a matter of some difficulty, owing to the inaccessibilityof certain enemy journals; and the same cause has resulted in theinclusion under the current year of some few papers which werenot available on account of their publication abroad in the latermonths of last year.It seemed better t o include these in the 1916Report rather than to pass them over in silence, althoughtechnically they belong to an earlier volume.During the year, the decline in certain branches of the subjecthas been marked. The study of plant colouring matters has appar-ently ceased.The chemistry of the blood and the bile, which usedto bulk largely in the Reports, has become practically stagnant.Chlorophyll seems to excite no interest a t present; and even thesupposed synthesis of the substance mentioned in last year’s Reporthas produced only one theoretical paper disagreeing with theauthors’ conclusions.1The chemistry of the alkaloids, once so fruitful a field, has beencomparatively neglected in the last twelve months, although thesynthesis of histidine by Pyman and the investigation of crypto-pine and protopine by Perkin stand out from a mass of otherwisesomewhat confused material. In this connexion, the Reportermay perhaps be permitted to hope that papers like these will bepreferred as models in place of the mass of ill-digested data andhasty conclusions which seem to form the stock-in-trade of certainl* W.Schlenk, D.R.-P., 292310 ; A., i, 683.l9 K. Lederer, Ber., 1915, 48, 1345, 1422, 1944, 2049 ; 1916, 49, 334, 345,1071, 1076, 1082; A., 1915, i, 1056; 1916, i, 40, 141, 208, 392, 646, 647.T. Jona, Mon. Xci., 1916, [v], 6, i, 149; A., i, 660.REP.-VOL. XIII. 130 ANNUAL REPORTS ON THE PROGRESS O F CHEMISTRY.foreign workers. Fully half the trouble entailed in reading paperson the alkaloids is attributable t o the fact that investigators hastent a publish a few isolated results as soon as these are obtained,instead of holding back the material until it falls into its properperspective among other data.Last year it was necessary to chronicle the discovery of newheterocyclic groupings in which fresh elements appeared asmembers of the rings.This year the field has been much enlargedby the introduction of silicon, arsenic, and titanium among theelements capable of producing heterocyclic compounds. On theother hand, the announcement of the discovery of silver pentazole,which was mentioned in last year’s Report, has turned out to beunfounded.The pyrrole group still retains its place as a centre of interest;but it will be noted t h a t the trend of investigation has passedaway from the naturally occurring pyrrole compounds and is tend-ing toward the purely synthetic types.The flavone and diflavone group has again been the object ofresearch, and much interesting material may yet be expected inthis branch of the subject.Taken as a whole, the heterocyclic series still retains the atten-tion of many workers; but the results of their investigations aremore diffused than in past years, so that an air of scrappiness isinseparable from any wide survey of the subject.Consciousnessof this defect was present in the mind of the Reporter in drawingup the following pages, and it is to be feared that readers willfeel as he himself did. I f this should unfortunately be the case,he trusts t h a t they will mingle sympathy with condemnation, forthe task of compiling the Report has been no light one.Some New Heterocyclic Types.U p to a very recent date, the heterocyclic compounds, althoughdiffering widely among themselves in character and complexity ofstructure, bore a certain family resemblance to one another owingto the fact t h a t the non-carbon atoms of the ring were drawnentirely from among three elements : oxygen, sulphur, andnitrogen.There might be various non-carbon atoms in the samering ; but, generally speaking, the chemistry of the heterocyclicseries consisted in ringing the changes on these three elements.Two years ago, the discovery of rings in which mercury atoms wereincluded threw a fresh light on the possibilities which were stillopen, and last year silicon, selenium, arsenic, and titanium werORGANIC CHEMISTRY. 131added to the list of elements capable of playing their part asmembers of cyclic chains.2Penta-, hexa-, andhepta-atomic ring systems have been prepared containing, in addi-tion to the usual carbon members, one of the elements phosphorus,antimony, bismuth, tin, lead, and thallium, whilst further synthesesof heterocyclic arsenic and silicon compounds have been d e ~ i s e d .~The general method of synthesis may be illustrated by a concreteexample. I n the preparation of phenylcyclopentamethylene-D hosD hineThe current year has expanded this group.as-dibromopentane is chosen as the starting point, and is firstconverted into the corresponding organo-magnesium derivative.The latter is allowed to react with the compound, PPhC12, inethereal solution. After evaporation of the solvent, the requiredsubstance is obtained by distillation under diminished pressure.When a five-membered chain is desired, a&dibromo- or dichloro-propane is substituted for the pentane derivative, whilst if adifferent element be needed as a ring member, it is inserted byusing the appropriate substance of the type RXHal,, in which Xis the element to be introduced and R is any organic univalentradicle.I n this way, six-membered cyclic derivatives of phosphorus,arsenic, antimony, and bismuth have been produced, which yieldthe usual " onium " compounds with ethyl iodide and with mercuricsalts.p-Tolylcyclopentamethylenephosphine, C,H,,P*C6H,Me, isdistinguished by giving an additive compound with carbon tetra-chloride, while ethylcyclopentamethylenebismuthine, C,H,,BiEt,is so readily oxidised that, if it is poured on paper, it inflamesspontaneously in the air.Turning to the behaviour of the corresponding five-memberedrings, it is found t h a t they can be synthesised even more readilythan the six-membered compounds.The phosphorus and arsenicatoms in these five-membered cyclic chains appear to be endowedwith specially active residual affinity, for the substances formadditive compounds with many solvents, inorganic haloids, etc.The simple cyclic nature of the new substances is shown by thefact that their molecular weights are normal and also by theirbehaviour on fission with chlorine, the parent chloro-paraffin beingSee Ann. Reports, 1914, 126 ; 1915, 129.G. Griittner and M. Wiernik, Ber., 1915, 4$, 1473 ; A., 1916, i, 92 ;G. Griittner and E.Krause, Bar., 1916, 49, 437 ; A . , i, 443.F 122 ANXUBL REPORTS ON THE PROGRESS OF CHEMISTRY.regenerated. The boiling points of the compounds agree withvon Braun’s rule, the replacement. of two ethyl groups by thepentamethylene ring having the effect of raising the boiling pointby approximately 50°.Similar results have been obtained by another worker in the caseof arsenic compounds.4 It is proposed that the nomenclature ofthe new series should be brought into line with t h a t of the pyridinegroup by terming the arsenic analogue of pyridine nrsedine, whilethe substance corresponding with piperidine would receive thename arsepedine. For example, methylcyclopentamethylenearsinewould be called l-methylarsepedine. It must be admitted t h a t ifmuch work is done in this field in the future, it will be necessaryto find some convenient short, substitute for the present somewhatcumbrous 11 om en cl a t ur e.The So-culled P e n t r i d e Compounds.I n last year’s Report5 a description was given of a series of sub-stances which were assumed t o contain a cyclic chain of fivenitrogen atoms.The class of compounds seemed of such interestt h a t it’ is with regret t h a t the writer feels constrained t o insertan extra adjective in the title of this section. It appears, how-ever, t h a t the observations of Lifschitz are erroneous, for the resultsof Curtius and his colleagues6 point to a conclusion different fromt h a t arrived a t by the earlier worker.Lifschitz, by treating cyanotetrazole with hydrazine hydrate,obtained a substance which he termed pentazolylacetohydrazidine,aiitl to which he ascribed the structure>N*CH,*C( :NH)-NH*NH,.y: NN:NCurtius and his collaborators observe, however, t h a t when cyanogenis passed into aqueous azoimide for the preparation of cyano-tetrazole, the latter substance is not the only product, for a vary-ing amount of a further condensation product is produced, namely,bistetrazole :N3H + C,N, + CHN,*CN;CHN4*CN + NH, + C,N,H,.When the reaction mixture is treated with hydrazine hydrate in4 E. IT. Zappi, Bull. SOC. chim., 1916, [ivj, 19, 151, 290 ; A., i, 575, 683.5 P. 142.T. Curtius, A. Darapsky, and E. Miiller, Ber., 1915,4$, 1614 ; A., 1916,1, 64ORGANIC CHEMISTRY. 133-absolute alcohol a t Oo, a white product separates out, which appearsto be the bishydrazonium salt of bistetrazole,This formula seems to be established by the fact that the com-pound yields benzaldazine when shaken with benzaldehyde.When the filtrate from this substance is boiled under reflux,ammonia is evolved and a mass of yellow crystals formed.Onwarming these with more hydrazine hydrate to complete the. reac-tion, there romains, after the ammonia has all been evolved, anorange-yellow mass which corresponds with Lifschitz’s pentazolederivative. The substance, however, appears to be actuallyditetrazyldihydrotetrazine,and an account is given of reactions corresponding with thisstructure.Although the compounds in question are extremely complicated,it seems beyond doubt that Curtius is right, and that Lifschitz didnot succeed in producing a pentazole derivative, so t h a t the searchf o r a five-membered ring containing only nitrogen atoms remainsopen to those who wish to investigate t h a t field134 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Friend points out that the three dissociable chlorine atoms aresupposed to hover vaguely about the nucleus.Such an hypothesishas certain advantages, owing to its indefiniteness, because it maybe interpreted to accord with almost any theory or suggested gradeof chemical combination,” but, 011 the other hand, it has its weak-nesses. For example, the “ co-ordination theory,” “ whilst assum-ing a definite valency for cobalt, nitrogen, and hydrogen atoms,denies an equally definite valency to chlorine unless, as in thecase of the pentainmine salt (11), it happens t o fall within thenucleus.” Friend also directs attention to the fact that the “co-ordination theory” ascribes to cobalt a valency of six, which isopposed t o ordinary experience.The idea of “ principal ” and“ auxiliary ” valencies does not get over the difficulty, since it nowseems to be admitted t h a t there is 110 real difference between t’hem.Thirdly, Friend points out that in the “ co-ordination theory ” thechlorine atoms directly attached to a nietallic atom are supposedto lose their ionising power-a postulate which is in direct contra-diction to ordinary chemical ideas.I n view of these difficulties, Friend advocates a return to strnc-tural conceptions of the metalammines, and his own views seem torequire much less straining of chemical conveiitions than is necessi-tated by the “co-ordination theory.”Taking the two substances mentionecl above, Friend assuines thatthey each contain a nuclear six-membered ring, iii the centre ofwhich, but not attached to it, lies the cobalt atom, held there byphysical forces only.The chlorine atoms directly attached to thecobalt atom possess their normal power of changing into ions, whilethe chlorine atom which forms part of the ring is not capable ofionisation. I n this scheme, all the atoms retain the valenciesknown to be associated with them according to ordinary chemicalideas, so that, there is no need for hypotheses as to “main” or“ auxiliary ” valencies.C1 H, C1 z3 / N /\ \/ ‘\\//.\/ /\// /\ /‘ /\H,K / NH, H3N / NH3 1 CO---~--- c1 I Co--l---- C1H3N \ NH3 H,N \ XH3c1 \/\N \H, C1( T .1 (IT.)A test case betweeii t h e two theories is supplied by tlie lieliaviouORGANIC CHEMISTRY. 135of potassium ferrocyanide. Briggs7a has shown t h a t this salt occursin two modifications, which he termed the a- and P-forms. Nowsince, on the ‘‘ co-ordination theory,” the potassium atoms arehovering uncertainly around the f errocyanide nucleus, no isomer-ism can be ascribed to their positions relative to one another; andsince the nucleus is symmetrical, on the ‘‘ co-ordination theory,”the arrangement of its constituents can afford no explanation ofthe isomerism.(Werner himself admits t h a t isomerism of theferro- and ferr -cyanides cannot be covered by his views.)On the other hand, Friend’s theory furnishes an explanation, notonly of the existence of two ferrocyanides, but also of the occur-rence of potassium ferricyaiiide in isomeric forms. There are threepossible structures for potassium ferrocyanide :\ KN C 7- KN’”[‘C uNKI Fe ! 1-c N \ c KN C ... gK-.,-CiN-... ... KN’ 1 c c Fe-N I Fe Ic \N ... 1: IC EK CKN ...\NKiC/ \NKiC/ \ N f C /Ortho-form. Meta-form. Para-form.Of these, the ortho-form is simply the double salt, 4KCN,Fe(CN),,as is indicated by the dotted line through the formula. The meta-and para-f orms represent the two isomeric f errocyanides. Briggshas shown t h a t the a-ferrocyanide is stable in alkaline solution, butpasses into the fi-forin in neutral or acid solvents.Now when anaxid solution of potassium ferrocyanide (&form stable) is treatedwith nitric oxide it yields nitroprussic acid, which exists in oneform only. An examination of the formulz of the meta- and para-structures will show t h a t this can only occur if the @-salt has asymmetrical structure-which proves it to be the para-form. Onthis basis, the structure of potassium nitroprusside is as shownbelow :Similar reasoning applies to the case of the isomeric ferri-Since the @-ferricyanide is stable in acid solution, it is cyanides.7a Briggs, T., 1911, 99. 1019136 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.derivable from the P-ferrocyanide, and hence the two salts arerepresented thus :KCN 1 CNK\CN/B-salt./-y-\ CNKKCN\CKN/a-salt.The identity of the Prussian-blues obtained from ferric salts anda ferrocyanide or from ferrous salts and a ferricyanide, is a matterwhich fails of sattisf actory explanation on the co-ordinationtheory”; but on Friend’s theory it is easily explicable.Theformulae below will make the matter clear:CNKCN’ 1 \CXKKCN 1 CNICFeCl, + 1 Fe 1\CN/B-Ferrocyanide.KKCN CNFeCI, + I Fe’lCN ’ ‘CN\CKN/,B-Ferricyanide.\ \Prussian - bIue.The existence of the three known soluble Russian-blues can alsobe accounted for by Friend’s theory.Another attempt to explain the nature of the ferro- and ferri-cyanides has been made by Denigks.8 On his scheme, potassiumferrocyanide is represenbd by (I) and ferrous ferrocyanide by(11), whilst the production of Williamson’s salt from potassiumferrocyanide at a red heat is shown in the equation below (111) :/C- N-C: N K /C==N-C: N,I (p) F/’ ~ : N ~ % N : c > F ~ (.)\ I I(1.1 (11.)F ~ ( &NK C:NKI \~=N-c:NK \C==N-C:N/8 G.DenigBs, Bull. SOC. chirn., 1916, [iv], 19, 79 ; A., i, 310ORGANIC CHEMISTRY. 137NK,C=N-C: NK C: N-G-JY : C2Fe/ k N K &:NK = 6KCN + Fe<&N-c--N.C I >Fe.\ I .. \C=N--& NK NK(111.)Other examples of the applicability of the author’s views are givenin the original paper. They differ to some extent from the resultsarrived a t by Friend, since in most of his formulae Denig6s regardsiron as bivalent, whereas Friend employs tervalent iron atomsalso.These papers indicate a welcome revival of structural ideas inthe region of the complex salts.The conceptions of structurewhich have been so laboriously and skilfully built up in onebranch of chemistry are probably equally applicable in freshfields, although their growth may be slower than it was among thepurely carboE compounds, and it is encouraging to note the successof the newer views in just those particular cases which the old“ co-ordination theory ” failed to explain.Friend’s theory has been criticised by Turner,g but the criticismfails on certain points. Turner considers that the retention of thecobalt atom by the “shell ” is “not convincing,” and he findsfaultl with Friend’s views on the ground that in their present formthey give no reason to assume optical activity in the case of cwtainmetalammines.H e objects also t o chlorine being treated as a ter-valent element, and he quotes Friend as referring “ t o tervalentchlorine as the usual form in which combined chlorine is to befound.” What Friend actually says in the passage referred to byT,urnes is that “ t h e chlorine atom in the ring (on the Friendtheory) is tervalent and saturated,” there being no reference tothe tervalency of the element being usual.” This rather detractsfrom the accuracy of the criticism. The possible tervalency ofchlorine can scarcely be a matter of dispute.Another line of argument employed by Turner concerns thestability of bridged linking in a six-membered ring; but as manybridged rings are known among the alkaloids and terpenes thepoint scarcely seems a strong one, even if it be granted that analogycan be stretched so far as t o include the peculiar type of ringassumed by Friend.A further criticism is based by Turner on an examination of the‘‘ positive ” and ‘‘ negative ” valencies exerted by nitrogen atoms ;but this argument has considerably lost its point, since the dis-covery of triphenylmethyltetramethylarnmonium,l0 in which all thelo W.Schlenk and J. Holtz, Ber., 1916 49, 603; A., i, 385.E. E. Turner, T., 1916, 109, 1130.F138 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.valencies of the nitrogen atom are saturated by hydrocarbonradicles.Turner states that it has been shown (although not yet pub-lished) that sodium a- and P-ferrocyanides are identical, and notisomeric.I f this be so, the extension of Friend’s theory t o coverthis case will be unnecessary. The accounts of the experimentalwork in the matter will be awaited with interest, as the subject isimportant; in the meantime, judgment must be reserved until theevidence is made known.Steric Hindrance and Reactivity.Although questions of steric hindrance do not now excite theinterest shown in them twenty years ago, when the subject wasnovel, yet scarcely a year passes without the addition to our know-ledge of some new facts in this region of chemistry. A t the sametime, our outlook on the problem has changed.Instead of thepurely mechanical hypothesis, which depended upon the assumptionof actual collisions of atoms or interference in space with theirfree passage to and fro, we are now apt to look for some deepercauses in the case of inhibitions which in earlier years would havebeen definitely ascribed to steric hindrance pure and simple.A very good example has come to light during the past year.11It has been shown that tertiary aromatic amines can be condensedwith formaldehyde to form para-substituted benzyl alcohols, and astudy has been made of the effect produced by substituents in thephenyl nucleus. A t the same time, the influence of the same sub-stituents on the capacity of the tertiary amino-group to yieldquaternary ammonium salts has been examined. The figures belowgive the relative yields in the two cases uiicler comparableconditions.Derivative of dimethylaniline o-Methyl o-Chloro- o-Rromo- o-Methosy-Yield of benzyl alcohol, per cent.6 36 45 GOYield of ammonium salt, percent. ... ... ... 7.6 15.6 16 100An examination shows that the two reactions can scarcely beinfluenced by steric hindrance to exactly the same extent. I n theone case, the reaction takes place a t the nitrogen atom, in theortho-position with respect t o the substituent, while in the secondreaction the active sphere is a t the other end of the benzenenucleus, greatly removed from the substituent so far as mere spatialconditions are concerned. Yet, as the figures show, both reactionsare affected in the same manner by the presence of the substituentJ.von Braun, Ber., 1916, 49, 1101 ; A . , i, 647ORGANIC CHEMISTRY. 139atoms. The only conclusion which call be drawn is that thehindrance in this case has its origin, not' in purely spatial sources,but rather in some arrangement of residual affinity within themolecule. The colour of the alcohols-they are yellow oils-tendst o confirm this view, for if both the alcoholic radicle and the amino-group retained their usual structure, there is no apparent reasonfor any development of visual colour in such compounds.I ntesting the reactions of formaldehyde and methyl iodide withkairoline (l-methyltetrahydroquinoline), the reactive hydrogenatom in the last compound was found to be much more readilydisplaced than might have been anticipated, whilst the reactivityof the nitrogen atom also was enhanced, in spite of the fact thatit has adjacent to it the second ring of the quinoline nucleus.Steric influences alone would have exactly the opposite effect.Somewhat similar conditions appear to come into play in thecase of 2 : 3-dimethyl-y-benzthiopyran7 which forms additive com-pounds with a large number of metallic salts.12 Although theevidence is somewhat scanty, it seems probable that intramolecularrearrangement accompanies the formation of these additivecompounds.Similar peculiarities mark the resuIts in some other cases.A ATew Indicator.Hitherto there has been no satisfactory indicator for hydrogen-ion concentrations lying between the limits of 6 and 8 on theSorensen scale, but this year the synthesis of dinitrobenzoylene-carbamide has filled the gap.I3 The substance has the structureshown below (IV); the first nitro-group lies in the position 6, butthe orientation of the second one has not yet been established.(111.) (TV.)The steps in the preparation of the new indicator are shown iuthe above formulae.Anthranilic acid (I) is treated with potassiumcyanate solution, yielding a precipitate of o-carbamidobenzoicacid (11). By adding sodium hydroxide and cooling, the sodiumsalt of the acid separates out, and when this is dissolved in boilingwater and acetic acid is added, benzoylenecarbamide (111) isl2 H. Simonis and A. Elias, Bcr., 1916, 49, 1116 ; A , i, (360.l3 M.T. Bogert arid G. Scatchard, J . Amer. Chem. SOC., 1916, 38, 1606 ;A . , i, 672.F* 140 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.obtained in a 92 per cent. yield. After nitration, this substancegives the required indicator (IV).The indicator is colourless a t the, limit 6 and is greenish-yellowat the limit 8. It is very slightly affected by neutral salts, canbe used a t looo or in the cold, and is not more affected by proteinsthan is p-nitrophenol. With ammonia and hydrochloric acid i tgives sharp end-points, but it cannot be used for the titratioiz ofcar b ona tes .The Pyrrole Group.Researches in this branch of the heterocyclic series continuewith almost unabated vigour, and in view of the importance ofthe pyrrole derivatives from the physiological side, any extensionof our knowledge in this field must be welcomed.I n previousyears, the members of the pyrrole group round which most con-troversy centred were those connected with the problem of thebile constituents, but during the past twelve months the mainsubjects of investigation have been substances of a much simplercharacter.To commence with the synthetic section of the subject, mentionmust be made of a modification of a method already known. Itwill be recalled that the interaction of amino-ketones and &ketonicesters gives rise t o pyrrole derivatives, and i t has now been shown l4that compounds containing a secondary amino-group can be usedin place of the amino-ketones without inhibiting the capacity forreaction.When phenacylaniline is boiled with either ethyl or methyl aceto-acetate, the product of the reaction appears to be 5-hydroxy-4-acetyl-1 : 3-diphenylpyrrole, which seems t o be formed by a con-densation expressible as follows :Pyrrole derivatives containing a substituted propyl group as aside-chain might become of interest if they could be applied t o asynthesis of alkaloids, such as tropine or hygrine, so that a recentlydevised method of preparing such members of the pyrrole seriesmay a t a later dats have an important bearing on the synthesisof naturally occurring substances.The method consists in allow-ing the potassium derivative of pyrrole to react with epichloro-hydrin, and the reaction appears t o take place in two stages.I nthe first stage, the potassio-pyrrole attaches itself t o the oxide14 G. K. Almstrom, Annalen, 1916, 411, 350 ; A., i, 568.18 K. Hess and H. Fink, Ber., 1916, 48, 1986 ; A., 1916, i, 168ORGANIC CHEMISTRY. 141ring, whilst the second stage entails the elimination of pot,assiumchloride :C,H,NK + CH,--CH*CH,Cl + C4H4N*CH2*CH(OK)*CH,Cl \/0 / YC,H,N*CH,*CH,---CH, + KCl\/0The reaction also gives rise t o a dimeride of the new ring compound.The reaction between potassio-pyrrole and dichloroisopropylacetate (acetyldichlorohydrin) is not quite so simple as this. Fromthe fact that the product contains a new diacetylpyrrole, i t seemsprobable that the hydrin decomposes into epichlorohydrin andacetyl chloride. The properties and reactions of the pyrrylpropylene oxide have been studied, and certain bicyclic pyrrolederivatives have thus been obtained.One of the products of the oxidation of pyrrole is a substanceknown as pyrrole-black; but along with this there are formed,according to the conditions of the experiment, other substanceshaving colours ranging from brown to yellowish-white.16 Severalof these compounds have been examined, and it appears that theircompositions are very similar t o that of pyrrole-black itself.Theyellowish-white substance appears to have a marked tendency tofurnish blue compounds. A study of the action of hydrogenperoxide in acetic acid solution has resulted in some results ofinterest, but the details cannot be given here.A new oxidation reaction17 has been worked out which appliesto the cases of some pyrrolidine and piperidine derivatives.Bymeans of formaldehyde, secondary hydrindamines of the pyrrol-idine group can be converted into tertiary amino-ketones :C,H,N-CH,*CHMe*OH -+ C,R7NMe.CH,-COR/Ie.When other aldehydes are used, the reaction gives rise t o l-alkyl-pyrrolidyl ketones ; and when primary alcohol derivatives are em-ployed in place of secondary alcohols, the resulting products arealdehydes.The reaction is rendered more interesting owing t o the markedinfluence exerted on i t by the solvent employed. I n alkaline solu-16 A. Angeli and L. Alessandri, Atti R. Accad. Lincei, 1916, [v], 25, i, 761 ;17 K. Hess, F. Merck, and C. Uibrig, Ber., 1915, 4.8, 1886; A . 1916,A., i, 667.i, 67142 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.tions the results are quite different from those obtained when anacid solveiit is used :C,H 7NMe- CH,* CH2*OHa k d i n C/+ C,H,N*CH,-CH,*OH + CH,Oacid \* C,H7NMe*CH,-CH0The exact processes which take place during the reaction are notyet quite clear; but it seems likely that there is an intermediateproduct of the aldehyde-ammonia type produced, so t h a t thestages in the case of pyrrolidyliwpropyl alcohol might be formu-lated as follows :**CH,*NH CH,*N*CH,*OHCH,/ 1 +CH,O -+ CH,/ I +\C H,*CH* CH (OH )Et \CH,*CH*C€?(OH)Et(1.1 /CH2*YHa WH2* CH*COEtCH,,(111.)The aldehyde-ammonia compound is formed( 11.1as shown in (11), andthen the group marked ** osidises the group marked it so as toproduce the substance (111).If a third reagent is present wliicli is more readily oxidisablethan the group marked with the single asterisk, this new reagentis oxidised instead and the alcoholic radicle is shielded. Thus inpresence of formic acid the reaction results in the formation ofcarbon dioxide aiid the compound :C H,*y,llecH,<,,2-c H m ( o H ) E tA curious reversal of the oxidation reaction is produced in somecases by the action of hydroxylamine on the end-product.Hydroxylamins appears to favour the addition of water to theketones, and a t the end of the reaction formaldehyde may bedetected in the solution.Tlie reverse reaction is supposed to takethe following course :An application of the reaction in a wider field would be oORGANIC CHEMISTRY.143interest. I n the case of piperidinel8 the action of formaldehydesolution leads to the formation of almost equal quantities ofmethylpiperidine and methylenepiperidine.The Con version of Pyrazoline Derivatives in t o cycloPara f i t ! s.It will be recalled that when pyrazolinecarboxylic acids areheated they decompose with the evolution of nitrogen and theformation of acids derived from the cycloparaffins. A n attempt l9has been made t o extend the application of t,his reaction into afresh field; and the results, although not quite so simple as hadbeen expected, are none the less of value, and are, perhaps, evenmore interesting than they would have been had they fallen outexactly as anticipated.Pyrazolines are obtained when liydrazine is allowed to react witha-olefinic a,ldehydes or ketones :CH,: CH*CHOC6H,*NH--NH,and it was assumedQ H,*CH,= fi HC,H,- N--- N 3that distyryl ketone would react in a mannersimilar to that which is observed in the case of simpler substances.It appears, however, t h a t the reaction is very much influenced bythe conditions of experiment.Thus when not more than one mole-cule of hydrazine is employed for each molecule of distyryl ketone,the end-product is 5-phenyl-3-styrylpyrazoline (I) :CHPh:CH*CO/H CH'CHPh ,imNH, -+ CHPh:CH*CO<NaNH, CH:CHPh ~CH,*$: H P hN--NHCHPh: CH* C<-(1.1If, on the other hand, hydrazine is employed in excess, the reactiontakes place as before, but in addition two molecules of the newlyformed pyrazoline combine with one molecule of hydrazine, givinga well-defined compound of this type :CHPh*NH*NH CHPhCH,*CQN CH,* hrr: C HPh I--(11.)The reaction obviously parallels t h a t which takes place betweenhydroxylamine and distyryl ketoxime, addition of the amino-com-pound taking place a t the remaining double linking of the ketone.K.Hess and F. Wissing, Ber., 1915, 48, 1907; A . , 1916, i, 74.l o N. Kishner, J . Russ. Phys. Chem. Soc., 1915, 47, 1819 ; A., 1916, i, 290144 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.These two pyrazoline derivatives behave differently when heated.The compound (I) when heated alone gives rise to 3:4-diphenyl-The compound (11), on the other cy clop en t ene,hand, suffers various types of decomposition according to the coii-ditions chosen.When hesatNed alone, i t breaks down in either oftwo ways : (1) it yields, in the first place, hydrazine and 5-phenyl-3-styrylpyrazoline (compound I), which then gives the cycloparaffinderivative already mentioneJd; o r (2) i t undergoes scission a t thehydrazine residue, giving rise t o 5-phenyl-3-fl-plienylethylpyrazoline(111), which, on further heating with potassium hydroxide, pro-duces l-phenyl-2-j3-phenylethylcycZopropane (IV) :YHPh-CHCHPh- CH2 >CH.(111. )I + CH,Ph*CH,= CH-CHPh\/CH,N2 +(IV. )(111.)CH,Ph*CR mCH-CHPh(IV. 14'\/C'H,N2 +When heated with hydrochloric acid the compound (11) gives thehydrochlorides of hydrazine and of the compound (I).This hydro-chloride of the compound (I), when further heated, suffers a lossof hydrogen chloride, and this decomposition is accompanied byintramolecular change, which gives rise t o the hydrochloride of adipyrazoline base, as shown below :CH-C-CH, CH2-C-CHz-+ l $ l PhCh---N--CH Ph li ! I PhCH NH-CHPh/\H C1/\H C1It will be seen from the above that even a comparatively simplereaction is capable of ramifying in many interesting directionsORGANIC CHEMISTRY. 145Some Sulphur Compozcnds.I n former years the reporter found i t possible to arrange a roughclassification of the researches which had been carried out in thisbranch of the heterocyclic series ; but during the past twelvemonththe investigations of this group have been so isolated and variedin their objects that it is almost impossible t o separate the papersinto definite categories. Thus the present section must necessarilyappear sornewhat scrappy in its form; the only division line whichcould be drawn was between compounds containing sulphur atomsalone in addition t o carbon ones as ring-members and other com-pounds which contain further elements in the ring in addition t osulphur ones.The simplest coinpounds requiring mention are the new sul-phones of the thiophen series, which have been obtained from theparent substances by the action of 30 per cent.hydrogen peroxidein glacial acetic acid solution.20 When tetraphenylthiophen istreated in this way i t yields tetraphenylthiophen dioxide,C,Ph,SO,.This substance appears to bO a tnie sulphone, anddiffers in character from the compound which results from theaction of hydrogen peroxide on thiophen itself. From the theo-retical point of view, the new sulphones are not without interest,as they mark a case in which the sulphur atom of thiophen isforced out of its usual inactivity, and their formation points t othe likelihood that the inertness of the thiophen sulphur atom withrespect to alkyl iodides is not ascribable to steric influences, butmust be sought in some chemical phenomenon.A rather unexpected result 21 has been observed when n-octaneis heated with sulphur in a sealed tube f o r some hours at 270-280O.A thiophen. CRHI2S, and a thiophthen, C,H,S, are formed, althoughin small yield.It is suggested that the first stage in the reactionis the conversion of the normal carbon chain into a branch-chainof the following type, CHMe2-CHMe.CHhle,, which is attacked bythe sulphur, with the formation of a compound having the follow-ing structure : v g M e 9 s--g MeCH C CH\/\/ 8 sThe action of sulphur on indene, hydrindene, and cyclopenta-A vigorous reaction ensues dime has led t o curious results.222o 0. Hinsberg, Ber., 1915, 4.8, 1611 ; A . , 1916, i, 66.W. Friedmann, Ber., 1916, 49, 1344 ; A . , i, 735.23 Ibid.? 50 ; A . , i, 415146 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.when iiidene and sulphur are heated together, and a compound,which appears to be a di-indenethiophen of the following formula,is produced, along with a second substancecorresponding with C,,H,,S :having a compositiona substance, CISH12S, The action of sulphur on hydrindene gavewhich was formed in an almost pure condition, whilst with dicyclo-pentadiene a simpler substance is produced, which has the composi-tion Cl,Hl,S.An int,eresting series of sulphur derivatives has been obtainedby the action of phosphorus pentasulphide on o-benzoic-sulphinide.23 The first product of the reaction appears t o be thio-o-benzoicsulphinide, which is then converted to some extent into2 : 3-dithiosulphindene :Better yields of this last compound are produced by acting on(( thiosaccharin " with phosphorus pent,asulphide :2 : 3-Dithiosulphindene, when heated with dilute alkali carbonateor hydroxide solution, is degraded t o an open-chain substa,iice, asshown in the following scheme:C,H,<-E>S + SH*C,H,*C02H + C0,H*C6H,.S.SoC6H,*C0,H.A new synthesis24 of l-thiochromones (benzthiopyrones) and1 : 4-dithiochromones (4-thiobenzthiopyrones) has been worked outas follows.Thiolbenzenes with free ortho-positions are condensedwith &ketonic esters in the presence of phosphoric oxide, and inthis way thiochromones are formed in the same manner as theoxygen analogues. When these thiochromones are melted withphosphorus pentasulphide, 1 : 4-dithiochromones are produced.Thus from thiophenol and methyl acetoacetate, 2 : S-diniethylbenz-thiopyrone is produced, which, on treatment with phosphoruspentasdphide, gives 4-thio-2 : 3-dimethylbenzthiopyrone,23 A.Rlannessier, CTazxetta, 1916, 46, i, 231 ; A., i, 231.z4 H. Simonis and A. Elias, Ber., 1916, 49, 768 ; A , , i, 499ORGANIC CHEMISTRY. 147Heterocyclic compounds containing nitrogen and sulphur atomsin their rings have been studied in some detail during the year.A very simple synthesis of benzthiazolines2s has been found to takeplace when o-aminothiophenolhydrochlorides are shaken in agueous-alcoholic solution with aldehydes. The reduction of the thiazolineto the corresponding thiazole has been noted when the thiazolineis heated with methyl iodide. Apparently the reaction followsthe course shown in the scheme below:C,H,<yE>CHPh + Me1 -+ C,H4<fE>CHPh I% - CH;;Other thiazole derivatives have been synthesised 2G by removingthe elements of water from the formyl derivatives of aminoethylor aminopropyl thiols :CH,--SCH29NHI CHO -+ - 11.20 I \CH.CH,*SHCH,-NH/A new class of indigoid dyes has been prepared.?J Hitherto,compounds of this type have been confined to those in which boththe groups a t the ends of the ring are cyclic in character, but thisyear a successful attempt has been made to synthesise aliphatic-aromatic indigoid dyes.The older dyes were prepared by theaction of compounds containing the group *CO*CH,* on the anilidesof diketones, b u t when it was desired to apply this reaction toopen-chain ketones, such as methyl ethyl ketone, i t was found thatthe ketomethylene group alone did not possess sufficient reactivityto permit the reaction to take place.Accordingly, a P-diketonewas substituted for the monoketone, with the result t h a t condensa-tion was found to occur quite readily. I n this manner, dyes wereformed which contained the chromophoric group *CO-C:C-CO-.Curiously enough, however, the new dyes have no affinity forvegetable or animal fibres.Some work on hydroxy- and methoxy-derivatives of “ thioindigo ”may also be mentioned here, although f o r details the reader isreferred to the original paper.2* The results prove that com-pounds of this class behave as normal dyes.Space forbids the summarisation of papers on the naphtha-z5 M. Claasz, Ber., 1916, 49, 1141 ; &4., i, 660.26 S. Gabriel, ibid., 1110 ; A , , i: 668.z7 1%’. Herzog and A. Jolles, ibid., 1015, 48, 1574; A.: 1916, i, 74.2 s P.Friedleender, ibid., 1916, 49, 955 ; A., i, 674I48 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.sultams 29 and the derivatives of sulphazone.30 These must beconsulted in the original. As to the oxidation of thiodiphenyl-amine by means of mercuric oxide or mercuric acetate,31 it willbe sufficient to mention that the product is a heterocyclic com-pound of the following structure :The conversion of phenazoiiium perbromide into the bromide ofmethylene-blue takes the following course :C IF eN-->C,H, + 4Br + ZNHRle, =G *'SBrThe reaction is stated to be very suitable for a lecture experi-ment.32TJbe C'OZLT)K~T~IL Group.0110 or two points of interest have come to light in this fieldduring the year.When coumarin is warmed with 20 per cent.sodium hydrogen sulphite the solution deposits sodium hydro-coumarinsulplionate,33 which crystallises with one molecule ofwater :Although solutions of the sodium salt are neutral, they can betitrated with sodium hydroxide. This peculiar behaviour seems tofind its explanation in a series of reactions. The first action ofthe sodium hydroxide is t o eliminate the sulphonic group with theformation of coumarin and sodium sulphite, whilst this is followedby a combination of sulphite and coumarin to produce a sulphonicderivative of hydrocouinaric acid :If the solution is evaporated with excess of alkali present, a cou-marate is formed, and this reaction provides a rapid method ofconverting coumarin into coumaric acid.On the other hand, ifz 9 T. Zincke and C. Julicher, Annulen, 1916, 411, 195 ; A., i, 426.30 M. Claasz, Ber., 1916, 49, 350; A., i, 424.31 L. Pesci, Gazzetta, 1916, 46, i, 103; A., i, 289.32 F. Kehrmann and R,. Speitel, Ber., 1916, 49, 63 ; A., i, 435.F. D. Dodge, J. Amer. Chem. Soc., 1916,38, 446; A., i, 413ORGANIC CHEMISTRY. 149the dry residue left on evaporating the solution is heated withace tic anhydride, the hydrocoumarinsulphonate is regenerated, andfrom it couniarin can be obtained by adding 20 per cent. sodiumcarbonate. Thus the following series of reversible changes can becarried out:Couinarin hydrocoumarinsulphonateliydrocouniaric-sulphonate coumaric acid.A study of the hydrolysis of coumarin leads t o the followingresults : I n alcoholic solution with a large excess of alkali, coumarinis completely hydrolysed.If then the alkali is partly neutralissd,coumarin begins t o be formed, even although the solution is stillmarkedly alkaline. By the formation of the coumarin, the solutiontends to become more alkaline again, and the process can be re-peated by the further neutralisation of the alkali up t o the pointwhen couinarin and its salts come into equilibrium.Some modifications in the experimental details of coumarin syn-theses are reported, which are said t o yield better results thanthose given by the ordinary nietliod.34A synthesis of thiocoumarin by distilling o-thiolcinnamic acidwith phosphoric oxide appears to produce good yields.35A peculiar example of autoxidation has been observed in thecoumarin series.36 A specimen of 1 : 4-dimethylcoumaran-2-one,which had been stored for some months in a corked bottle, wasfound to have become converted almost entirely into 4-acetoxy-m-toluic acid.It is supposed that this change is due to the additionof oxygen t o the ethylenic linking of the enolic form of thecoumaranone : 9- - C,H3Me@yl>CMe + 0, = C,H,Me< -C(OH)*? I0. CMe---OAs might be expected, the reaction was much more rapid when thecompound was freely exposed to air.Broniination appears to lead to analogous results. Thus when aclilorof orm solution of dimethylcoumaranone is treated with itsown weight of bromine a t Oo, bromo-pcresotic acid,separates out.HO*C6H2MeBr CO,H,34 H.Simonis and Goldenzweig, Ber., 1915, 48, 1583 ; A., 1916, i, 57.36 H. Simonis and A. Elias, ibid., 1916, 49, 763 ; A., i, 408.36 K. von Auwers, ibid., 820 ; A., i, 496150 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.I3 12 t i s. O ~ J 1.0 t ) P P ( I 1 ) d F7u L' o H e D ey i IW ti 2% e s.I n last year's Report37 reference was made to the blue iodineadsorption compounds formed by certain pyrone derivatives, andthis year has added a certain amount to our knowledge in t h a tfield. It is found 38 t h a t 8-phenyl-y-benzopyrone-2-carboxylic aciddissolved in alkali and poured into N / 100-potassium tri-iodide solu-tion containing excess of acid gives an intense blue coloration,whilst an alcoholic solution of the ester shows a similar behaviour.Even more sensitive in character is 8-phenyl-y-benzopyrone itself,since it yields a blue adsorption compound with iodine up to adilution of AT, 4000.A 1 per cent. alcoholic solution of 7-phen-anthropyroiie gave blue tints with aqueous iodine solutions up ton-jzO00. It appears from tlie results obtained up to tlie presentthat the presence of benzene nuclei in the pyrone molecules has amarked effect 011 tlie formation of the blue compounds.A modification of the flavone synthesis has been devised,39 inwliicli iiitriles (instead of esters) of acetoacetic acids are condensedwith phenols. The iiiiiiies thus formed can be liydrolysed toA avo ii es . Tliu s w lie n ace t y 1 p lie ii y 1 ace t oni t r il e is condensed wit liresorcinol and sulpliuriccourse :acid, tiie reaction takes tlie followingSO,HII NMeThe phenols wliicli reacthydroxyl substituents.best are those which contain methyl or-A study of the effect of substituents on reactions has been carriedout in the flavonol synthesis from benzylidenecoumaraliones."OHalogen-additive compounds of benzylidenecoulnaranones are de-composed by alkalis in either of two ways, as indicated below :3 i Ann.Reports, 1915, 144.3 8 A. R. Watson, T., 1916, 109, 303 ; A . , i, 414.39 B. N. Ghosh, ibid., 105; A . , i , 281.4 0 K. von Auwers, Ber., 1916, 49, 809; A., i, 406ORGANIC CHEMISTRY. 151The relative amounts of the two end-products formed is governedby the nature of the substituents introduced into the molecule.Thus ortho- or para-substituents in the benzene ring with respectt o the oxygen atom of the coumaranone ring considerably weakenthe reaction (11), so that the chief products of the reaction in thesecircumstances are flavonols, whilst meta-substituents have the oppo-site effect.Substituents in the phenyl ring of the benzylidenenucleus also affect the decomposition. In fact, tlie presence in thatring of a nitro-group or two rnetlioxy-radicles is sufficient t o iiihibitthe formation of any flavonol.T h e DifZavoiie c'~01cp.I n last year's Report 41 attention was directed to the synthesis ofcompounds derived from the hitlmrto unknown nucleus, diflavone :0 0/\/\/\II I I I1\/\/\/ co coAlthough these substances are highly coloured, they are not capableof acting as mordant dyes, owing to the absence of auxochromicgroups, and this year attempts have been made to prepare otherderivatives of the series which contain such radicles, and whichwill therefore be comparable with the dyes of tlie mono-flavonecompounds.In the course of this in~estigation,~2 certain points ofinterest have arisen which serve to ilIustrate the marked influenceexerted on the progress of reactions by the conditions under whichthey are carried out.The condensation of 4 : 6-diacetyl-l : 3-dihydroxybenzene withanisaldehyde was tried in four different ways. In the first place,the reaction was carried out in boiling alcoholic solution in the pres-ence of sodium hydroxide, and in this case the end-product was4 : 6-di-p-methoxycinnainoyl-1 : 3-dihydroxybenzene (I).Secondly,an aqueous solution of the diacetyl derivative, anisaldehyde, andsodium hydroxide was allowed to remain at, the ordinary tempera-41 Ann. Report, 1915, 154.42 H. Ryan and J. Algar, Proc. Roy. Irish Acad., 1916, 32 [ B ] , 185; A . ,i, 662 ; H. Ryan and M. J. Walsh, Proc. Roy. Irish Acad., 1916, 32 [ B ] , 193 ;A . , i, 663152 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRYture for six months, whereby a stereoisomeride of the previous end-product was produced. Thirdly, the diacetyl derivative and anis-aldehyde were condensed in h o t alcoholic solution i n presence ofhydrogen chloride, and in this way a diflavanone derivative (11)was obtained after some heating. Finally, if the same operation isconducted a t the ordinary temperature, the product is the diflavan-one derivative (111) :MeO/'\OMeMeO~C6H,-CH: CH*CO~ 1 COGH:CH*C,H,.OM~ \/KO/\/\CH*C,H,.OM~MeO*C,H,* CH :CH* CO' I IC:CH*C,H,*OMe(11.10 0MeO*C,H4*CH/\/\l/)C H *C,H,*OMeCO CO(111.)When either of the compounds (I) is treated with sodium acetateand acetic anhydride it yields a diacetate.The two diacetates ontreatment with bromine in chloroform solution give tetrabromides,and these react with alcoholic potassium hydroxide with the forma-tion of the same crystalline derivative, which appelars to be adicoumaranone (IV) :\/\/GOMeO*C6H,-CH:C' C : CH C, 8,. OMe \/\/\/0 0MeO*C,H,*CH:C i/ 1 C:CH*C6H,*OMeco co(IV. 1When veratraldehyde was substituted for anisaldehyde, and the/\/\/'\\/\/\/first of the methods mentioned above was chosen, the result was asubstance (V) which, on subsequent acetylation and treatment withbromine and alcoholic potassium hydroxide, yielded a dicoumaran-one derivative (VII) :OMe/ CO-CH:CH*<I)OMORGANIC CHEMISTRY.153OMe)C:CH-C,H,( OMe),./-O-\/\/O-\\CO/\/\C*/(Me0)2C,H,* CH: C< I 1(VII. )The similarity between the new compound (V) and the curcumindimethyl ether, the structure of which is shown in (VI), is obviousa t a glance, and an examination was made of the comparativetinctorial properties of the two substances. The dicoumaranonehas no effect on unmordanted wool, whilst with tin and aluminiummordants it gave pale canary and pale lemon colours, curcuminunder the same condition yielding orange-red and orange-yellowtints.It is thus clear that the new substance has tinctorial effectswhich are slight in comparison with those of curcumin.Further research in t,his group will be awaited with interest, asi t is clear that the compounds are very sensitive t o changes in thereaction conditions, and much valuable information as to the effectof slight alterations in reagents may be expected.The Indole Group.A new indole synthesis43 has been discovered, based on theelimination of carbon dioxide and water from o-aldehydophenyl-glycine under the action of acetic anhydride and sodium acetate,whilst a series of photochemical syntheses of indole derivatives 44has been worked out.Autoxidation of indoles in the presence of sunlight45 has yieldedsome interesting substances, one of which appears to be an etherealderivative of indole having the following structure :A study of the reduction of certain indole derivatives by meansof sodium amalgam has shown that various end-products are43 W.Gluud, D.R.-P., 275282 ; A., i, 288.44 P. Pfsiffer, Annalen, 1916, 411, 72 ; A . , i, 327.45 B. Oddo, Gazzettn, 1916, 46, i, 323 ; A., i, 502154 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.formed according to the conditions of the experiment.46 Thus,1 : 1-dimethyldihydroindolium chloride (I) is converted by thereducing agent into three other bases, of which (11) is formed tothe extent of 75 per cent. of the total yield, (111) to the extentof 8 per cent., whilst (IV) makes up the remaining 17 per cent.I I I (Ti!/ (I f i ti, ( I t 1 t l if s Z l c t* i i ' ( I ti LT c .c.A rapid method of atmospheric oxidation of indigotin andsimilar substances has been devised,-'7 which depends o n exposingthem to air in solutions of compounds of high boiling point, suchas quinoline or phenanthrene. Thus, in boiling phenanthrene solu-tion, 1 gram of indigotin can be completely oxidised in half anhour.It seems possible that this method might be applied to othercases of oxidation.Another new method of oxidation48 depends on the utilisationof triethylphosphine as an oxygen carrier. The phosphine is con-verted into an unstable phosphine peroxide, PEt,:O,, which thenparts with some of its oxygen to the indigotin, and is itself reducedto the corresponding oxide.The reaction can be utilised as alecture experiment.Substituted indigotins can be produced49 by the reduction ofsubstituted chloronitroacetophenones by boiling with zinc and30 per cent. acetic acid. The reaction follows the course shownbelow :The Pyridithe Group.Some syntheses of pyridine derivatives have been worked outduring the last year. Thus, the condensation of ethyl sodio-malonate with the sesquihydrochloride of hydrocyanic acid gives a50 per cent. yield of ethyl 2 : 6-diketodinicotinate,50 in accordancewith the following scheme:46 J. von Braun and L. Neumann, Ber., 1916, 49, 1283 ; A., i, 742.47 P. Friedlaender and N. Roschdestwensky, ibid., 1915,48, 1841 ; A ., i, 80.4 8 N. Kjshner, J . Russ. Phys. Chem. Soc., 1916, 47, 2129; A., i, 290.4 9 F. Bodinus, Chem. Zeit., 1916, 4Q, 326 ; A., i, 429.5 0 L, Gatterninnn and -4. Sltita, B e y . , 1916, 49, 49.1; A . , i, 419ORGBKIC CHEMISTRY. 155coC0,Et NH, C0,EtI C0,Et NH*II I I I2CHNa + Cl-C-Cl + &H--CH-bH + 2NaC1 +C0,Et H C0,Et C0,Et,CHco \g/The ester is readily convertible into dihydroxypyridine, from whicha number of derivatives have been produced, of which the mostinteresting are azo-dyes correspondiiig with the formula/\ N,XHO‘ ]OHN\/Pyridine bases are also obtainable by the condelisation ofketones with amides.51 Thus, when acetone (2 mols.) and acetamide(1 mol.) are heated together in a sealed tube to 250°, 2 : 4 : 6-tri-niethylpyridine is formed.2 : 4 : 6-Triphenylpyridine is formedsimilarly from benzaniide and acetophenone. I n neither case, how-ever, is the yield a good one. From analogy, it might be expectedthat. pyridine itself could be produced by this reaction from acet-aldehyde and formaldehyde, but experiments in this direction havebeen unsuccessful. a-Picoline is formed in sinall yield fromparacetaldehyde and acetamide. Various pyridine derivatives s2have been isolated by hydrolysing casein in presence of methylal,the latter compound being utilised as a permanent source of form-aldehyde. I n this reaction, primary, secondary, and tertiary basesare produced, but quinoline derivatives do not seem to be formed.Reference was made in last year’s Report53 to the action ofsodium on pyridine.Since then, further work has been done inthis field.54 The first product of the interaction between sodiumand pyridine is the substance, (C,H,N),Na, and it has been shownthat, on heating a t 130°, this is converted into the simpler sub-stance, C5H,NNa, by loss of 1 molecule of pyridine. Analogousresults are obtained when sodium is replaced by lithium orpotassium, two compounds of each metal being formed. It seemspossible t h a t a further investigation of this problem may throwinteresting light on some questions of valency.A somewhat peculiar behaviour has been observed in the addi-61 A. Pictet and P. Stehelin, Compt. w z d . , 1916, 162, 876 ; .4., i, 671.52 A. Pictet and T. &. Chou, ibid., 127 ; A., i, 226.53 Ann.Report, 1915, 141.54 B. Emmert, Ber., 1916, 49, 1060; A . , i, 454156 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.tion of amyl nitrite t o pyridine.55 When the two substances aremixed in a sealed tube, no interaction appears to occur even aftersome weeks, but if the tube is opened, crystals appear in a day,which prove to be pyridine nitrate. When amyl nitrate andpyridine are allowed to remain together, no reaction takes place.It is possible that the reaction can be explained by the decomposi-tion of the alkyl nitrite in the presence of moisture, and that thenitrous acid thus set free attaches itself to the pyridine and thenundergoes oxidation, but it is also possible that the pyridine decom-poses the nitrite, with the formation of amylene and nitrous acid.A peculiar catalytic action of thionyl chloride has been observedin some applications of the Friedel-Crafts’ reaction to the pyridineseries.56 When absolutely pure materials are employed, no reac-tion takes place between pyridine or quinoline derivatives andacetyl o r benzoyl chloride in the presence of aluminium chloride.A similar inhibition is noted when aromatic hydrocarbons andpyridinecarboxylic acid chlorides are employed.In neither caseis any ketone produced. I f , however, a trace of thionyl chlorideis added t o the mixture, the reaction proceeds readily along thenormal lines, and ketones are produced in good yields. The meresttrace of thionyl chloride seems sufficient to start the reaction.Ifthe acid chlorides have been prepared by the action of thionylchloride on the acids, it is found that after exhaustion a t theordinary temperature enough thionyl chloride remains behind t ostart the process of ketone synthesis.The Quinoliae Gropp.A new and simpler synthesis of isoquinolines has been devised.57Aromatic aldehydes are condensed with acetone ; the products arereduced t o saturated ketones ; syn-oximes of these are prepared ;and finally the Beckmann rearrangement is brought into play.Three products are thus obtained; (I) the acetyl derivativeof the amine; (11) the methylamide of the acid; and (111) a1-methyl-3 : 4-dihydroisoquiiioline.RGHO -+ R*CH*CH*COMe + R-CH,-CH,*COMe +R-CH,*CH,*CMe:N*OHI V*.A”:N(11- 1 (IIT.) (1.1b5 C. W. Addy and A. I<. Macbcth, T., 1916,109, 754 ; A . , i, 668.66 R. Wolffenstein and F. Hartwich, Be?-., 1915, 48, 2043 ; A., i, 222.5’ A. Kaufmanii and R. RadoseviE, ibid., 1916, 49, 675 ; A., i, 502ORGANIC CHEMISTRY. 157By proper selection of the experimental conditions, it is possible toprocure a good yield of (111).Some oxidation 58 and reduction 59 reactions of quinoline deriv-atives have been studied, but i t is unnecessary to give detailshere.Mercuric Acetate as a Reagent in Alhaloiclcrl Chemistry.The use of mercuric acetate as a mild oxidising agent has longbeen known, but recently60 it has been shown that this substancecan be utilised, not only qualitatively, but quantitatively, since itis possible t o recover and weigh the sparingly soluble mercurousacetate which results from the reaction.Not only so, but owingt o the mildness of the reagent's action, it is sometimes possiblet o conduct the reaction in stages, which are quitc clearly defined.Thus, for example, when N-methylbulbocapnine is oxidised bymeans of alcoholic iodine, it loses two hydrogen atoms, whilst athird is replaced by iodine. When mercuric acetate is employed,the result is a simple removal of two hydrogen atoms from thecompound if the reaction is carried out a t the ordinary tempera-ture, but on the water-bath a further stage of oxidation takesplace and an extra molecule of mercuric acetate is used up. I n asimilar manner, corydaline is converted first into a dihydro-corydaline and then into a tetrahydro-derivative.Other examplesgiven are the oxidations of d-canadine, 1.-laudanosine, and papa-verine. It seems that the reagent will prove of great value infurther investigations.Some New Blkcrloid Syntheses.The constitution of damascenine (froni Niglelln damnscem) wasestablished by Ewins61 in 1912, and this year a new method ofsynthesis62 has been discovered which enables the alkaloid to beprocured in quantity. The starting material is commercial8-hydroxyquinoline, which, by the action of methyl sulphate andpotassium hydroxide, is converted into 8-methoxyquinoline.Further treatment with methyl sulphate produces an additivecompound having the structureR. Boehm and K. Bournot, Ber., 1915,48, 1570 ; A., 1916, i, 75.6g J.von Braun and E. Aust, ibid., 1916, 49, 501 ; A., i, 421.6o J . Gadamer, Arch. Phtim.., 1915, 253, 274 ; A., 1916, i, 736.G1 Emins, T., 1912, 101, 544.G2 A. Kaufmann, E. Rothlen, and B. Vargolici, Rer., 1916,449, 578 ; A., i,417158 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Oxidation of this last substance, by means of aqueous potassiumpermanganate in the presence of magnesium sulphate, gives riseto a mixture of methoxy-X-niethylisatin and formyldamasceninicacid, MeO.C,H,(NMe*CHO)*CO,H. The latter compound, ontreatment with 5 per cent. methyl-alcoholic hydrogen chloride, isconverted into damascenine itself (methyl 2-methylamino-3-meth-oxybenzoate). If f ormyldamasceninic acid is treated with dilutehydrochloric acid, it is hydrolysed to formic acid and damasceninicacid, whilst the action of concentrated hydrochloric acid leads tothe formation of the hydrochloride of dainasceninic acid.A new synthesis of histidine has been accomplished in the follow-ing manner.63 The 4-hydroxymethyl derivative of glyoxaline ( I )is prepared in the usual way, and is oxidised with chromic acid,yielding glyoxaliiief ormaldeliyde (11).The latter compound isthen condensed with hippuric acid by means of acetic anhydride,and the result is 2-phenyl-4-[l-acetylglyoxaline-4-methylidene]-oxazoloiie (111). When this oxazolone is boiled with very diluteaqueous sodium carbonate, the acetyl group is removed and theoxazoloiie ring opens, and if the calculated quantity of an acid isthen added, the result is the formation of a-benzoylamino-b-glyoxaline-kacrylic acid (IV), which can be reduced to benzoyl-r-histidine by the usual methods. Histidine (V) is finally obtainedby hydrolysis.>CH.TH2 SH-NH$!H*CH2*C---NC0,H(V. 1The yields obtained in the various reactions are satisfactory.TI1 e JIorpll ine G T O U ~ .A considerable amount of investigation during the year hascentred on the study of morphine alkaloids in which the N-methylgroup is replaced by cyanogeii or hyclrogen.G4 The preparation ofsuch compounds has been made easier by utilising the action of achloroform solution of cyanogen chloride or bromide on a similar63 F.1;. Pymnn, T., 1916, 109, 186; A . , i, 335.G' J. von Braun, Her., 1916, 49, 750 ; A . , i, 500 ; ibid., 977 ; A., i, 665ORGANIC CHEMISTRY.159solution of the acetylated alkaloid.65 From the cyano-derivatives,ether may be obtained by the action of sodium ethoxide and analkyl iodide. The physiological action of inany of these morphinederivatives has been st~..died.~4The reduction of thebaine and phenyldihydrothebaine 66 hasbeen carried out in various ways with the object, of determiningwhether or not thebaine contains ethylenic linkings in its structure,as Knorr suggested in his formula (I). The results are regardedas disproving the presence of double bonds in the molecule, andthe formula (11) has been suggested as more in accordance withthe evidence./\OMe1 llCH(1- 1By the action of methylal on-\ ? / '(11.1tetrahydropapaverine,G7 a single\/CH,base- is produced (in contradistinction from the action of acetal,which gives rise to two coralydines).The new compound is R,lower homologue of the coralydines, and has been named nor-coralydine. It has the structure shown in the formula below.CH,I lOMeOMe\/Papaverine has the property of condensing with aldehydes (orsubstances capable of giving rise to aldehydes), and this powerseems t o be shared by many of the papaverine derivatives68F. Hoffmann, La Roche & Co., J3.R.-P., 289273 ; A . , i, 417.66 M. Freund and E. Speycr, Ber., 1916, 49, 1287 ; A . , i, 738.6 7 A. Pictet and T. Q. Chou, ibid., 370; A., i, 418.6 8 Soci6t6 pour 1'Industrie Chiniique 21 RBle, Fr. Past,., 475654 ; A . , i, 221160 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Further work has been done on berberine derivatives. Azo-derivatives of the dihydroberberine series have been prepared,69the azo-group replacing the reactive hydrogen atom in the posi-tion 4.The demethylation of isoberberine70 has been accomplished,and i t is now assumed thatfollowing structures :OH CH-,CHzI /\I I\/this substance has one or other-c: H,CH2Iof theCryptopine n,nd Protopine.An important paper7I has been published on these substancesand their derivatives. Fortunately, it appeared in the Trans-actions, and is thus easy of access, f o r it is clearly impossible tosummarise even the salient points in the space a t the Reporter'sdisposal. All that can be done is to set down some of the moreimportant matters with which it is concerned, and leave thereader to examine the data and arguments in the original.Cryptopine is an alkaloid occurring in opium and having thecomposition C21H2305N.It is optically inactive and has a markedphysiological action.Perkin, from a study of the various decompositions of thealkaloid, draws the conclusion t,hat its behaviour is b,est representedby the formulaCrypt opine.He points out, however, that this formula does not account for the6 9 M. Freund and K. Fleischer, Annalen, 1916, 411, i ; A., i, 325.7O IT. Scholtz, Arch. Pharm., 1915, 253, 622 ; A . , 1916, i, 416.71 W. H. Perkin, jun., T., 1916, 109, 815ORGANIC CHEMISTRY. 161fact that cryptopine yields neither oxime nor semicarbazone,although the structure given above contains the group *CO*CH,~,nor does the alkaloid react with isoamyl nitrite and sodiumethoxide, as might be expected from a structure such as this.Further, the ten-membered ring is a somewhat unusual grouping,although, since the synthesis of nonamethylene, there seems t o beno reason for rejecting a formula on that ground alone.Thestructures of the methylcryptopines are also considered.Various reactions of cryptopine have been studied-nitration,oxidation with different reagents, and also sundry decompositionsof the alkaloid. A very full account of the salts of cryptopine isgiven.Protopine, C,,H190,N, is a very widely distributed alkaloidwhich occurs along with cryptopine in opium. The two alkaloidsdiffer in composition by one carbon and four hydrogen atoms, andsince they both give many similar reactions, i t seems probable thatthey are closely akin to one another. Cryptopine (see above) con-tains two methoxy-groups which are not found in protopine, andit appears probable that these two groups are replaced, in thelatter alkaloid, by a single methylenedioxy-radicle. Perkin hasbrought a large amount of evidence to show that the structure ofprotopine may reasonably be regarded as this:CO NMeCH2<g/\’ CH, I\/‘CH,/Protopine.It is impossible t o go further into detail, as the evidence is com-plex and does not lend itself to summarisation, but the reader isadvised to consult the original, which extends to more than twohundred pages.Y o h.i m b in e .The nature of this interesting substance has given rise t o a con-siderable amount of controversy. I n last year’s Report72 it wasassumed that the identity of yohimbine and quebrachine had beenestablished, as seemed t o be demonstrated by the work of Fourneau72 Ann. Report, 1915, 163.REP.-VOL. XTII162 ANISUAL REPORTS ON THE PROGRESS OF CHEMISTRY.and Page.73 Spiegel,74 however, refuses to accept this evidence.It appears that Fourneau and Page compared yohimbine with aspecimen of I ‘ quebrachine ” furnished by Merck, but Spiegel main-tains that this preparation is really a mixture of yohimbine andmesoyohimbine.As to the constitution of yohimbine, Barger and Field75 agreedwith Spiegel that the composition of the parent alkaloid was repre-sented by C,,H,,O,N,, but they believed that yohimboaic acidcould best be regarded as C,,H2,0,N,, whereas Spiegel76 contendsthat yohimbine is the N-methyl ester of yohimboaic acid, andshould therefore have the composition C,oH,403N,. He rests hisdissension from Barger and Field’s conclusions on the ground thatthe barium salt chosen by them for analysis was not a suitablesubstance for examination.Further light has been thrown on the constitution by a studyof an intermediate compound formed from yohimbine and capableof conversion into yohimboaic acid.76 This substance is producedwhen yohimbine is heated with half a molecule of dilute alcoholicpotassium hydroxide. Further hydrolysis converts it intoyohimboaic acid. This intermediate substance occurs free intechnical yohimbine, from which it can be isolated by crystallisa-tion from 50 per cent. alcohol. It is dextrorotatory, gives acrystalline hydrochloride, and contains a methoxyl group.Attempts t o apply the Hofmann reaction t o yohimbine havebeen U ~ S U C C ~ S S ~ U ~ , ~ ~ but the results are not without interest. Whenyohimbine methiodide is treated with potassium hydroxide, methyl-amine is liberated, whilst with silver oxide the methiodide isoxidised if the reaction is carried out in the presence of light. I nthe dark, silver oxide acts on the methiodide in cold methyl-alcoholic solution, with the production of methylyohimboaic acid.From this it follows that the methyl group in this last acid isattached to a nitrogen atom which, in yohimbine itself, is free, andin this way it is clear why attempts t o prepare yohimbine byesterifying the acid have always proved fruitless.The results of investigations of the constitution of yohimbinehave not thrown light on its actual structure as yet, but the datawhich are being accumulated certainly give cause for hope that this73 E. Fourneau and H. J. Page, B d l . S c i . Pharmncol., 1914,21, 7 ; A . , 1914,74 L. Spiegel, Ber., 1915, 48, 2084 ; -4., 1916, i, 287.‘5 G. Barper and (Miss) E. Field, T., 1915, 107, 1025.76 L. Spiegel, Bey., 1915, 4$, 2077; A., 1916, i , 286.7 7 L. Spiegel and IT. Corell, ibid., 1916, 49, 1086 ; A . , i, 667.i , 862ORGANIC CHEMISTRY. 163valuable compound may at no very far-off date be within the scopeof synthetic chemistry.Scopoliii e .This alkaloid has been the subject of three papers during the lasttwelve months, but, as seems customary in the case of alkaloidchemistry, one of these is concerned chiefly with claims as t opriority. The other two papers deal with the subject fromdifferent points of view, and apparently they support each other.It appears probable that the constitution of scopoline may bsrepresented by the formula C7Hg,NMe,0H,0:, which, after all,does not lead us very far. An attempt78 has been made t o throwsome light on the group C,Hg. The oxidation of scopoline itselfdid not seem to lead far in this direction, but since scopoline canbe quantitatively converted into norscopoline 79 by potassium per-manganate, the nor-compound was used instead of the parent sub-stance. An attempt t o utilise the new method of oxidation bymeans of formaldehyde (see p. 141) failed, as norscopoline simplyregenerates scopoline when submitted to this process. The deduc-tion drawn from this was that scopoline contains a tertiaryhydroxyl radicle.Scopoline was therefore heated in a sealed tube with hydrogenbromide in glacial acetic acid, and in this way a hydrobromidewas formed which is easily reduced t o a hydroscopoline. Oxida-tion of the latter compound produced 1-methylpiperidine-2 : 6-dicarboxylic acid (scopolic acid). This reaction gives the positionsof all the carbon atoms in scopoline, and proves that the alkaloidis a piperidine derivative having the skeleton shown below:C-CScopolyl chloride also yielded negative results.The proof of the constitution of scopolic acid was obtained bythe synthesis of the compound in the following manner.80 Lutidinewas oxidised with potassium permanganate to lutidinic acid,which was then reduced by the Paal-Skita method (hydrogen andcolloidal platinum) to hcxahydro-2 : 6-lutidinic acid. The hydro-chloride of this, when heated with formaldehyde solution a t 135cfor four hours, gave 1-methylpiperidine-2 : 6-dicarboxylic acid.7 8 K. Hess and A. Suchier, Ber., 1925, 444, 2057 ; A . , 1916, i, 285.7 g Ibid., 1886 ; A., 1916, i, 67.80 K. Hess and F. Wissing, ibid., 1907 ; A . , 1916, i, 74.a 164 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.c H2OCO,H*CH CH*C02H -+-/\ -H2 SH2A /\3\/CH 1' ICH, -+ C02H(>0,H -+N \ / NCO,W*CH CH*CO,HN Me\/Another worker81 has arrived a t similar conclusions on thescopoline question, and suggests that the formula for hydro-scopoline may beCH,-CH-CH-OHI Y I ICH,-CH--CH*OHHydroscopoline ?If these views should prove correct, the elucidation of the scopolineconstitution is almost achieved.For the disconnected nature of the present survey the Reporterhas already offered his apologies, but, in conclusion, he feels itdesirable t o point out that this disconnectedness has had itsadvantages also, for it reproduces to some extent the impressionwhich is derived from a study of the literature of the year, andso it serves to mirror the general trend of things in this branchof chemistry better, perhaps, than could have been done by a moredetailed survey of restricted fields. Reports of the progress inthe Heterocyclic Division are almost necessarily less interestingthan those which deal with the other two sections of the subject.I n the aliphatic and homocyclic series, test cases are found fornew theories as well as for the study of chemical constitution, butin the heterocyclic group the main interest is confined t o theinvestigations of molecular structure and the application of newmethods of synthesis or degradation. Thus one of the mostinteresting features of organic chemistry is, generally speaking,absent from this Report, although in the current year it has beenpossible to deal with a theoretical problem in connexion withFriend's views on the metalammines.A. W. STEWART.E. Schmidt,, Arch. Pharm., 1916, 253, 497; A . , i, 2 8 5 ; see also Rer.,1916, 49, 164 ; A . , i, 419

ISSN:0365-6217    

DOI:10.1039/AR9161300066

出版商:RSC    

年代:1916

数据来源: RSC

摘要:

ANALYTICAL CHEMISTRY.AN immediate, and to some extent unforeseen, consequence of thewar was the failure of a supply of glass suitable for laboratoryvessels and of paper of the necessary quality for filter-paper foranalytical purposes. Both difficulties have now been overcome.The formulz €or manufacturing glass of the right composition forevery description of chemical apparatus were worked out last year,]and filter-papers of all types are now obtainable. The quality ofthe British filter-paper is quite equal, and in some cases superior,to that of its predecessor, for the necessity of supplying this wanthas caused special attention t o be directed to the physical charac-teristics of good filter-paper, and the nature and origin of the im-purities likely to be present.2 For a moderately rapid paper, whatis technically known as the “bulk,” or, in other words, the ratioof the volume of the paper t o that of the fibre, should be about3.5 t o 1, and as a rule i t varies from 3-4.5 t o 1.The occurrenceof pinholes, which are sometimes to be found in paper of this bulk,is attributable t o faults in the milling.Physical Methods.Increasing attention is being given to the utilisation of the physi-cal characters of chemical substances as proofs of identity, andmethods which until comparatively recently were only used empiri-cally in restricted directions are now being more generally em-ployed in analytical work.This is particularly the case with the determination of the vis-cosity, which was for many years confined to the examination ofoils by means of standard instruments.No special precautionswere taken to eliminate the numerous sources of error, and theresults obtained with one kind of instrument were not comparablewith those obtained with another.The adoption of a general method of expressing the viscosityof a liquid in absolute units would obviate these drawbacks, andwith this end in view various formulae have been worked out forProc. Inst. Chern., 1915, ii, 33.E. J. Bevan and W. Bacon, Anolysl, 1916, 40, 159.10166 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.calculating the measurements given by one standard instrumentinto those of another, and f o r expressing the results in terms ofabsolute viscosity.For example, Ostwald’s viscometer has been standardised bydetermining the efflux velocity of sulphuric acid from the capillarytubes a t 25O.3 The specific gravity and the percentage strength ofthe acid gave the required data for obtaining the viscosity by refer-ence to the graph.4 The absolute viscosity (11) was calculated bymeans of the formula 71 = Kdt, where d represented tlie specificgravity, t tlie efflux velocity in seconds, and I< a constant depend-ing on the size of the capillary tube of the particular instrument.I n the case of Redwood’s viscometer it w7as found that the relation-ship between the efflux velocity and the absolute viscosity could beexpressed by means of a straight-line graph.It is a general practice in commercial work to ignore the specificgravity of the liquid in viscosity measurements, but this may leadto considerable error.5 The results may be corrected by multiply-ing the viscosity measurement by the specific gravity of the liquida t the temperature of the determination compared with t h a t ofwater a t 4O.Another source of error is the erroneous assumptiont h a t the average pressure in the instrument is exactly the mean ofthe initial and final pressures. To obviate this, a method of calcu-lating tlie true average pressure has been devised.6 The necessarycorrection t o be applied for the loss of kinetic energy when a liquidissues from a capillary tube of uniform diameter, with its ends a tright angles to its axis, has been ascertained, b u t i t has yet to bedetermined whether the same correction is applicable to a capillarytube with a funnel-shaped opening.Pending this investigation, itis advisable not tlo use the latter form of tube.7In view of the practical inconvenience attending the use of thepresent unit of absolute viscosity, the question of the advisabilityof employing a unit of lower value has again been raised. Thereis much t o be said in favour of the proposal t o term the viscosityunit the ‘ I poise ” (after Poisseuille), with the ‘ I centipoise ”( l c p = 0 . 0 1 ~ ) as the practical unit of measurement. This valuewould be practically the same as the specific gravity of water a t20°, and on this basis the viscosity of an oil expressed in centipoiseswould be the same as its specific viscosity a t t h a t temperature.8C. A.Savill and A. W. Cox, J . SOC. Chem. Ind., 1016,35, 151.A. E. Dunstan, P., 1914, 30, 105.P. C . McIlhiney. J . I n d . Eng. Chem., 3916, 8, 433.6 E. C. Bingham, H. I. Schlesinger and A. B. Coleman, J . Ainer. Chern.SOC., 1916, 38, 27 ; -4., ii, 321.E. C. Bingham, H. J. Xchlesinger, and A. B Coleman, Zoc. cit.P. C. McIlhiney, Zoc. citANALYTICAL CHEMISTRY. 167The chief contribution to the applications of the spectroscope inanalytical work has been an investigation of the differences in thespectra of the alkali salts of the phthaleins of different phenols.These show distinctive absorption bands, the different positions ofwhich *afford the means of identifying the various phenols.9Reference may also be directed t o an important paper dealingwith the methods of measuring absorption spectra and comparingthe wave-lengths with those of a standard pure substance of knowncomposition.10The measurement of the optical dispersion has now become astandard method for testing the purity of Chinese wood oil.Themethod has been simplified by placing the oil in a hollow prism inthe spectrometer and taking a direct reading of the dispersion bymeans of tin arc light. The relative dispersion for two lines givenby this light in the red and blue parts of the spectrum is deter-mined, and from the difference between the angular measurementsgiven by Chinese wood oil and its principal adulterants, the degreeof purity of the sample may be rapidly ascertained.11Gas Analysis.Tlie systematic study of the different reagents used or suggestedfor use in the analysis of mixtures of gases has been continued fromlast year,12 especially by American chemists, and several papersgiving details of considerable value have been published.Some of the reagents which have, in recent years, been proposedfor the separation of individual constituents of gaseous mixtureshave little to recommend them beyond their novelty, whilst othershave not fulfilled the expectations of their proposers.For example,the use of a solution of phosphorus in castor oil, which Cent-nerszwer13 suggested f o r the absorption of oxygen, is quite unsuit-able for the purp0se.1~ At the ordinary temperature the absorptionof the gas is incomplete, which may possibly be due to the oil onlyretaining a small amount of phosphorus in solution.It is conceiv-able that by tlie use of a hot solution of pliosphorus in oil completeabsorption of oxygen might bs effected, but such a reagent wouldnot be altogether easy to manipulate, and even assuming that,under such conditions, i t gave quantitative results, i t would not9 H. Gsell, Zeitsch,. anat. Chem., 1916, 55, 417 ; A., ii, 584.10 F. Weigert, Ber., 1916, 49, 1496 ; A . , ii, 545.l1 E. E. Ware, J . Ind. Eng. Chem., 1916, 8, 126.l2 R. P. Anderson, J . Ind. Eng. Chem., 1915,7, 587; A . , 1915, ii, 647.13 M. Centmerszwer, Chem. Zeit., 1910, 34, 494; A., 1910, ii, 541.14 R. P. Anderson and W. Biederman, J . Ind. Eng. Chem., 1916, 8, 136;A., ii, 262168 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.offer any advantage over the ordinary method of absorption withalkaline pyrogallol.Chromous chloride solution is a much more promising reagent,but so far it, too, has failed t o justify the claim that it couldadvantageously replace alkaline pyrogallol.15 Unfortunately, solu-tions of chromous chloride prepared in the usual way by the inter-action of chromous acetate and hydrochloric acid are unstable, andgradually decompose, with the liberation of hydrogen, thus render-ing analytical results untrustworthy.On the other hand, a solu-tion of chromous chloride prepared by reducing violet chromicchlorids with hydrogen a t 400-500° and filtering the liquidthrough glass wool, is a stable reagent, but does not absorb oxygencompletely.Although the conditions f o r preparing a satisfactorystable solution might be established, the cost of the materials andthe trouble of preparing i t would probably prevent its general adop-tion in gas analysis, notwithstanding the fact that the presence ofcarbon dioxide o r hydrogen sulphide does not interfere with theabsorption of oxygen by chromous chloride.For use with a viscous solution of potassium pyrogallol, whichhas been found a more efficient absorption reagent than weakersolutions,lG Orsat’s original pipette gives better results than theother forms of pipette devised as improvements on t h a t instrument,for the absorption of oxygen by weak solutions, without shaking.It is quite effective f o r the absorption of oxygen from the air, butfor the analysis of gases rich in oxygen, shaking is required, and amodification of Orsat’s pipette has been designed to preventobstructions caused by the formation of precipitates.18 The modifiedpipette has a conical top, and the glass tubes are supported on acone of perforated porcelain.Solutions of sodium pyrogallol have been found to absorboxygen more rapidly than any solution of potassium pyrogallol,complete absorption being effected within four minutes.19 A solu-tion which has five times the absorptive capacity of Anderson’s mosteffective potassium pyrogallol reagent 20 consists of sodium hydr-oxide, pyrogallol, and water in proportions of 7.36 : 10.0 : 11.62 byweight.A solntion of this description is too concentrated for anycarbon monoxide t o be liberated on contact with oxygen, althoughl5 R.P. Anderson and J. Riffs, J. I n d . Eng. Ckem., 1016, 8, 54 ; -4.,l6 R. P. Anderson, ibid., 1915, 7, 587 ; A.. 1915, ii, 647.l7 R. P. Anderson, ibid., 1916, 8, 131 ; A . , ii, 262.l a R. P. Anderson, ibid., 133 ; A . , ii, 262.l9 J. W. Shipley, J. Amer. Chem. Soc., 1916, 38, 1687 ;2o R. P. Anderson, J. Ind. Eng. Chem., 1915, 7, 587; A . , 1915, ii, 647.ii, 261.-I., ii, 571ANALYTICAL CHEMISTRY. 169it has been established 21 that dilute solutions of sodium hydroxidewill evolve carbon monoxide under these conditions.Attention has been directed 22 to a source of error in the analysisof generator gas through incomplete absorption of carbon monoxideby ammoniacal cuprous chloride, and the combustion method istherefore more trustworthy for the estimation of hydrogen,methane, and carbon monoxide in such gases.The method of frac-tional combustion with copper oxide23 has been shown to yieldaccurate results with gaseous mixtures of all kinds.24 The mostsuitable temperature for the combustion is 275-300°.As direct methods, when practicable, are more satisfactory thanindirect methods, a new process of estimating carbon monoxide inthe presence of unsaturated hydrocarbons seems worthy of thoroughinvestigation.25 It is based on the fact that when a gaseous mixtureis repeatedly passed through a tube charged with soda-lime (con-taining about 20 per cent. of sodium hydroxide) heated a t 230°, anycarbon monoxide present is absorbed and retained by the soda-lime.Obviously, the gas must first be freed from carbon dioxide andoxygen by being passed through absorption pipettes charged withthe usual reagents.The method appears t o be capable of generalapplication, and should further investigation confirm its trust-worthiness, it should find a permanent place in schemes for theanalysis of gases.A rapid and accurate process of estimating sulphur in small quan-tities of coal gas has been based on its oxidation by combustionwith an excess of air in a quartz tube containing platinum in theforin of gauze and wire fabric.26 The most efficient reagent f o rabsorbing sulphur compounds in gas is phenylhydrazine. The com-bustion test in the presence of porous platinum may also be usedfor the detection of sulphur in 10 C.C.of gas.A colorimetric method of estimating acetylene has beendevised by two chemists, apparently working independently of eachother.27 Essentially it consists in passing the gas through anammoniacal solution of cuprous chloride t o which gelatin has beenadded to prevent the subsidence of the resulting cuprous acetylide.The red colour produced is compared with standard colours, whichV. B. Lewes, J Xoc. Chena. Ind., 1892, 10, 407.* 2 F. Hoffmann, Chem. Zeit., 1916, MI, 412 ; A., ii, 395.23 G. von Knorre, ibid., 1909, 33, 717; A . , 1909, ii, 698.24 G. A. Burrell a.nd G. G. Oberfell, J . Ind. Erzq. Clmn., 1916, 8, 228 ;25 A. Piva, Ann. Chirn.Applicata, 1916, 5, 82 ; A . , ii, 343.26 F. Mylius and C . Huttner, Ber., 1916, 49, 1428 ; A., ii, 57127 E. R. Weaver, J . Amer. Chern. SOC., 1916, 38, 352; d., ii, 2 7 5 ; A.A . , ii, 260.Schulze, Zeitsch. angew. Chenb., 1916, 29, i, 341 ; A . , ii, 649.G170 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.may be prepared from a solution of a red dyestuff suitably diluted,or may consist of standard red glasses. The method is capable ofdetecting 0.03 mg. of acetylene and of estimating quantities up t o2.0 mg., but i t is essential that, prior to the test, the gas shouldbe freed from hydrogen sulphide or notable quantities of oxygenor carbon dioxide.I n connexion with the subject of gas analysis, reference may bemade to a form of apparatus designed for the examination ofminute quantities of gases.28Agrict~ltural Analysis.The limitation in the supply of phosphates, due to the war, hascaused special attention to be paid to the methods of estimatingthe value of phosphatic manures for agricultural purposes, and itis not surprising that criticisms should have been directed againstthe methods of estimation commonly employed.I n the early part of the year a paper was published29 givingdetails of experiments on mineral phosphates of various origin,and it was shown that from 90 to 100 per cent.of their phos-phoric acid could be dissolved by 2 per cent. citric acid solutionprovided that a sufficient number of extractions were made, Freelime or calcium carbonate causes a pronounced reduction in thesolubility-so much so that in the presence of much calciumcarbonate the citric acid test as ordinarily applied is a measureof the calcium rather than of the phosphoric acid. On the otherhand, the greater the proportion of calcium in combination withthe phosphoric acid the greater is the solubility.The results ofthe test are also affected by the degree of fineness to which themineral has been ground, whilst calcining progressively reduces thesolubility. From the results of these experiments it would seemthat the citric acid test must be regardeck as of no value as acriterion of the relative suitability of phosphatic fertilisers forplant life, and that there is no reason why mineral phosphatesshould not give as good results as basic slag in agriculture.Even when used for the evaluation of basic slag, the citric acidmethod may yield fallacious results if the slag has been producedby the use of fluorspar.30 I n the case of such slags there is n:,definite ratio between the amounts of silica and dissolved phos-phoric acid, whereas in the case of slag readily soluble in citricacid the ratio between the two constituents is fairly constant.2 * P. A.Guye and F. E. E. Germann, J. Chirrt. phys., 1916, 14, 195; A . ,ii, 445.29 G. S. Robertson, J . SOC. Chew. Id., 1916, 35, 217; A., ii, 19630 Ibid., 216 ; A., ii, 186ANALPTICAL CHEMISTRY. 171IIence it would appear that in fluor spar slags the silica is not incombination with the phosphoric acid. Such slags resemblemineral phosphates in yielding the whole of their phosphoric acidto citric acid solution if sufficient time be allowed f o r solution, buttheir behaviour affords a further proof that the citric acid testcannot be trusted as a criterion of the agricultural value ofphosphates.The test has become so firmly established, however, that it isunlikely to be discarded, a t any rate for the analysis of ordinarybasic slags.I n fact, several modifications of the method havebeen described during the past year, most of them having as theirmain object the more efficient prevention of the interference ofsilica when present in large amounts.It has been shown31 that in the iron citrate method32 there isalways sufficient iron present in the iron citrate, even when pre-pared from ferric chloride poor in iron, to prevent the inter-ference of silica.The addition of too great a quantity of hydrogenperoxide causes the results to be too low, possibly owing to theinfluence of the silica. Both ferric chloride solution and hydrogenperoxide should have been recently prepared, and the reagentsshould be added in the following order: (1) iron citrate;(2) hydrogen peroxide ; and (3) magnesium oxide.33I n a modification of the citro-uranium method,34 the phosphoricacid is precipitated as ammonium magnesium phosphate,35 theprecipitate dissolved in acetic acid, and the solution titrated withuranium acetate, with potassium ferrocyanide as indicator. Theinfluence of silica is inhibited by boiling the solution with ferroussulphate and filtering it from the resulting precipitate.A new method of estimating the phosphoric acid insoluble incitric acid solution has been based on the establishment of adefinite relationship between the amounts of phosphoric acid pre-cipitated by ammonia and the citrate-insoluble phosphoric acid.By dividing the amount of phosphoric acid in the precipitate bythe factor 1.5, the amount of phosphoric acid insoluble in citricacid solution is obtained.36F o r the estimation of phosphorus in soil several good methodsare available, and comparative experiments have shown that con-31 M.Popp, Chem. Zeit., 1916, 40, 257, ii, A ; ii, 266.32 Ibid., 1912, 36, 937 ; A . , 1912, ii, 092 ; Landw. Versuchs-&at., 1913, 79,229, 465; A ., 1913, ii, 306, 876 ; Chem. Zeit., 1914, 38, 741 ; A., 1914, ii, 576.33 Celichowski and F. Pilz, Zeitsch. Zandw. Vers-Wesen Oesterr., 1915, 18,581 ; A., ii, 342.34 D. Crispo and R. W. Tuinzing, Landw. Versuchs-Stat., 1916, 88, 131 ;.4., ii, 342.36 C. H. Hunt, J . Ind. Eng. Chem., 1916, 8, 251 ; A., ii, 265.3i Popp, zoc. cit.a* 172 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.cordant results may be obtained either by fusion with alkalis, byWashington’s method of treatment with hydrofluoric and nitricacids, or by Fisher’s aqua regia The presence oftitanium o r tungsten in the soil does not affect the accuracy ofthe results, but vanadium should be eliminated by a method suchas that of Cain and Tucker.38 It is necessary to take into accoLmt,however, that treatment with hydrofluoric acid or ammoniumfluoride may cause a loss, on subsequent ignition, of phosphorus asphosphorus fluoride.The loss is slight in the case of alkali nietals,but considerable with phosphates of alkaline earth metals.39A rapid method of estimating calcium oxide in peat soils consistsin treating the soil with aqua regia, evaporating the mixture t odryness, extracting the residue with dilute acid, and precipitatingthe calcium as oxalate, after previous separation of iron andaluminium. The precipitate is washed, treated with concentratedsulphuric acid, and titrated with standard permanganate. Themethod is not applicable to soils containing much mineral matter.4oTest estimations of the carbonates in soil by absorbing theevolved carbon dioxide in barium hydroxide, as in Marr’s method,*land titrating the resulting barium carbonate, as described byhave shown that the method may be accepted as trust-worthy.43Organic ,4 nalysis.Qualitative.-It is only to a limited extent that i t has beenfound possible to establish group reagents in organic chemistry,for in many cases a reaction can only be regarded as distinctiveof a certain class of compounds when other classes are known tobe absent.For example, the blue coloration with characteristic spectrumwhich “ ninhydrin ” gives with amino-acids 44 is only specific whencertain conditions as to temperature and concentration areobserved.All ammonium salts give the same reaction in a slightlyalkaline solution,45 and hence other bases which yield ammonia orare readily oxidisable might be mistaken for amino-acids.46Z i W.0. Robinson, J . I n d . Eng. Qhenz., 1916, 8, 148 ; A . , ii, 265.38 J. R. Cain and F. H. Tucker, ibid., 1913, 5, 647; A., 1913, ii, 875.39 A. Davis and J. A. Prescott, J . Agric. Sci., 1916, 8, 136.40 R. A. Gortner, Soil Sci., 1916, 1, 505 ; A . , ii, 449.41 F. S. Marr, J . Agric. Sci., 1909, 3, 155 ; A . , 1909, ii, 938.42 J. R. Cain, J . I n d . Enq. Chewt., 1914, 6, 465; A., 1914, ii, 577.43 C. J. Schollenberger, ibid., 1916, 8, 427 ; -4., ii, 395.44 S. Ruhemann, T., 1910, 97, 2025.45 V. J. Ilarding and F. H. S. Warneford, J . Riol. Chenz., 1916, 25, 319;46 V. J. Hardiiig and R. M. MacLean, ibid., 337 ; A., ii, 459.A., ii, 459ANALYTICAL CHEMISTRY. 173Again, a general reaction for alkaloids containing a phenolicgroup (morphine, etc.) has been based on the coloration whichmost phenols give with a solution of titanium trioxide in presenceof sulphuric acid,47 the colour ranging from blood-red t o orangein the case of different alkaloids. Here, too, the distinctive natureof the test is weakened by the fact that proteins containing atyrosine group give a similar coloration.**A useful test for distinguishing between aromatic and aliphaticaldehydes has been based on the different behaviour of the classesof compounds on treatment with a chloroform solution ofacenaphthene in the presence of sulphuric acid. Aromaticaldehydes yield a green ring, changing to reddish-violet, a t thezone of separation, or, if the liquid be shaken, it becomes first greenand then reddish-violet, whilst aliphatic aldehydes yield black ordark green condensation products.The colour reaction, which isvery sensitive, is also given by aldoses and carbohydrates whichyield formaldehyde or a similar aldehyde when treated with coldsulphur ic acid .49A new reaction for the detection of picric acid is based on theformation of 2-bromo-4 : 6-dinitrophenol. This gives a red colora-tion with ammonia vapour or potassium cyanide, and the test isfairly sensitive. being capable of detecting 5 mg. of picric acid ina litre of beer or ~ r i n e . 5 ~A study of the various methods of detecting hydrocyanic acidhas shown t h a t the guaiacum-copper sulphate test is capable ofdetecting 0.00039 mg.of cyanogen in 10 C.C. of solution, whilstthe silver cyanide test and the method of converting the cyanogeninto thiocyanate are not sensit,ive to less than ten times t h a tamount.51A test for glycerol depending on the coloration given by itsvapours with a reagent consisting of magenta, sulphurous acid, andsulphuric acid52 has now been modified so as to be capable ofdetecting glycerides in the presence of hydrocarbons, waxes, etc.Under the prescribed conditions, the vapours from glycerides givea red coloration, changing t o blue on heating, whilst the vapoursfrom resins, waxes, etc., give a red coloration, which becomescolourless when heated.5347 Compare 0. Hauser and A.Lewite; Ber., 1912, 45, 2480; A . , 1912,4 8 G. DenigBs, Bull. SOC. chim., 1916, [iv], 19, 308 : A., ii, 544.4 9 R. de Fazi, Qazzetta, 1916, 46, i, 334 ; A , , ii, 457.8 O J. Castets, J . Phann. Chim., 1916, [vii], 13, 46 ; A., ii, 100.n1 G. Anderson, Zeitsch. anal. Chem., 1916, 55, 459 ; A., ii, 586.52 M. Frangois and E. Boismenu, J. Phamn. Chim., 1915, [vii], 11, 49; A . ,i, 847.1915, ii, 110. 63 M. Frangois, ibid., 1916, [vii], 13, 65 ; A,, ii, 155574 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.A test for detecting formic acid in acetic acid and vinegar,which is based on its reducing action on chromic acid,54 suffersfrom the same drawbacks as similar tests previously suggested.Reduction is caused not only by formic acid, but also by otherconstituents normally present in wine and grain vinegars, andthe test cannot be regarded as absolutely conclusive even in thecase of commercial acetic acid.An illustration of the way in which a colour reaction may bemisleading is afforded by the fact that Japanese soja beans con-tain a constituent which gives a violet coloration with ferricchloride.55 This might easily be mistaken for salicylic acid, but,unlike that acid, it does not give a coloration with Millon’s orJorissen’s reagents.The last-named reagent should therefore beused in testing for salicylic acid in beans.Of the various tests proposed for distinguishing between gumarabic and other gums and dextrins, the reaction given by basiclead acetate is the most distinctive. As confirmatory tests, acopper reagent prepared by mixing copper sulphate with sodiumhydroxide, and a reagent consisting of a mixture of ferric chlorideand alcohol, may be used.56Quantitative.-There have been few contributions to themethods of elementary analysis during the past year.The mostimportant, perhaps, is the adaptation of the chromic acid methodof combustion57 to the simultaneous estimation of carbon andhalogens in organic compounds, the modification being also applic-able to the estimation of carbon alone in substances the combustionof which is difficult by the ordinary method.58To effect complete combustion of traces of carbon monoxide, asilica tube containing platinised asbestos is connected with theoutlet of the reaction flask, whilst oxides of sulphur are absorbedin a U-tube packed with glass wool moistened with sulphuric acid.During the combustion, a current of oxygen is passed through theapparatus.When the method is used for the estimation of carbonby itself, t,he resulting carbon dioxide may be estimated eithergravimetrically or volumetrically. I n the latter case, the gas isabsorbed by standard baryta solution, the excess of which is sub-sequently titrated with acid. F o r the simultaneous estimation ofcarbon and bromine, the gaseous products are absorbed in tubescontaining standard sodium hydroxide solution containing a little64 P. Szebhyi, Zeitsch. Nahr. Genussm., 1916, 31, 16 ; A., ii, 542.55 H. C. Brill, Philippine J. Sci., 1916, 11, 81.56 C. E. Waters and J.B. Tuttle, J. Ind. Eng. Chem., 1916, 8, 413 ; A . , ii,57 P. W. Robertson, T., 1915, 107, 902.6 8 Ibid., 1916, 109, 215 ; A . , ii, 267.400ANALYTICAL CHEMISTRY. 175standard sodium sulphite solution, and the contents subsequentlytreated with barium nitrate and filtered. The filtrate is titratedwith standard nitric acid, and the bromine estimated volumetric-ally by the silver nitrate method. To obtain the amount of carbondioxide, the value of the bromine is deducted from the total alkali-metric result. I n the case of chlorine, a second heated silica tubeis required to decompose any chromyl chloride and retain thechromium as chromium oxide. Estimation of nitrogen byKjeldahl’s method in the residue in the reaction flask gives un-satisfactory results.A special form of combustion tube has been devised for thesimultaneous estimation of carbon, hydrogen, and mercury inorganic compounds,59 the mercury being absorbed and weighed ina small inner tube containing gold leaf or a spiral of gold wire.The inlet of this tube is packed with calcined asbestos into aconstricted portion of the combustion tube, and the escape ofmercury vapour or water through the joints is prevented by theintroduction of dry oxygen through a side tubulure in theouter tube a t a pressure somewhat higher than that within thetube.The chromic acid method of oxidation has also been founduseful for estimating the carbon in the organic non-sugar con-stituents of sugar scums.60Oxalic acid is frequently employed as an original standard inalkalimetry, and provided it is purified by recrystallisation, givestrustworthy results.For its titration with alkalis, with methyl-orange as indicator, it is advisable to add an equivalent quantityof calcium chloride before completing the neutralisation, so as t ocounteract the tendency t o the formation of hydrogen alkalioxalates.61 The iodometric method of Sander 62 is less accuratethan titration with alkali.63 Iodine is liberated with extremeslowness towards the end of the reaction, whilst the addition ofcalcium chloride causes the results to be too low.61 For accuracy,the gravimetric method of precipitating the oxalic acid as calciumcrxalate, and weighing the resulting calcium carbonate, is t o bepreferred to any volumetric process.63A method of estimating alcohol in the presence of phenolconsists in rendering the mixture strongly alkaline with sodiumhydroxide, distilling part of the liquid, destroying any phenol in6 9 V.Grignard and A. Abelmann, Bull. SOC. chim., 1916, [iv], 19, 2 5 ;6 O V. StanEk, Zeitsch. Zuckerind. Bcihm., 1916, 40, 201 ; -4., ii, 267.61 G. Bruhns, Zeitsch. anal. Chem., 1916, 55, 23 ; A . , ii, 158.G2 A. Sander, Zeitsch. angew. Chem., 1914, 27, 192 ; A . , 1914, ii, 482.63 A. Blanchetiere, Bull. SOC. chim., 1916, [ivl, 19, 300 ; A., ii, 543.A , , ii, 149176 ANNUAL REPORTS O X TEE PROGRESS OF CHEMlSTRYthe distillate by means of bromine, the excess of which is removedby thiosulphate solution, and distilling off the alcohol.64Messinger’s method of estimating acetone in presence of ethylalcohol,65 which depends on its conversion into iodoform, has thedrawback that iodoform is also produced from the alcohol in aproportion increasing with the temperature.This may be obviatedby using baryta or lime-water t o replace the potassium hydroxide.Under the s-pecified conditions only a slight error, for which a cor-rection may be made, is then introduced by the alcohol in mix-tures containing 10 per cent. of acetone, but in the case of mix-tures containing only about 1 per cent. the results are less trust-worthy, since the necessary correction for the alcohol is too largein proportion t o the acetone.66For the estimation of paracetaldehyde and acetal in the presenceof each other, advantage may be taken of the fact that they aredecomposed a t different rates when heated with diLute acids.67Acetal in 1 per cent.solution is rapidly hydrolysed when boiledwith a strong acid of a concentration of N/6000, whereas paracet-aldehyde is stable until the concentration is increased to N/2000,and is only rapidly hydrolysed when the concentration reachesN / 1 0 . I n practice it is preferable t o use an equivalent quantityof acetic acid in place of a strong acid. Acetaldehyde, if alsopresent, is separated by a preliminary distillation, whilst the acet-aldehyde resulting from the decomposition of the paracetaldehydeand acetal is distilled after each acid treatment. For its estimationeither Ripper’s hydrogen sulphite method or the neutral sulphitemethod of Seyewetz gives good results.A convenient method of estimating ferro- and ferri-cyanides intl10 presence of cyanides and thiocyanates is based on the titrationof the f errocyanide, with titanium trichloride, after the additionof a large excess of ammonium thiocyanate.Alkalinity due t ocyanides is neutralised with standard acid, while ferrocyanides areoxidised to ferricyanides by means of acid permanganate, or, ifthiocyariates are also present, by means of iodine solution. ThediRerence between the results of the titration with titanium tri-chloride before and after the oxidation corresponds with the ferro-cyanide present. Soluble chlorides and sulpliates do not interferewith the estimation unless present in combination with metals whichform insoluble double ferrocyanides.6864 J.Ehrlich, J . I?zcl. Eng. Ghem., 1916, 8, 240 ; A . , ii, 349.65 J. Messinger, Ber., 1888, 21, 3366; A., 1889, ii, 313.66 J. Rakshit, Analyst, 1916, 41, 2 4 5 ; A . , ii, 544.67 K. J. P. Orton niid (Miss) P. V. McKie, T., 1916, 109, 184 ; *4., ii, 384.6 8 F. G. W. Knapinan and E. L. Randall, Chenz. News, 1916, 113, 265;A., ii, 501ANALYTICAL CHEMISTIZY. 177Aoother application of titanous chloride is in the estimation ofsubstituted phenylhydrazines and p-nitrophenylhydrazones, whichare quantitatively reduced by boiling with the reagent, whilst phenyl-hydrazine is unaffected. The excess of titanous chloride is subse-quently titrated with standard crystal-scarlet solution.69I n the analysis of commercial benzols the toluene may be esti-mated by distilling the sample a t a specified rate and measuringthe quantity of distillate obtained up t o 90°.Thence by referenceto a curve the amount of toluene, within about 1 per cent., isobtained. Carbon disulphide is estimated from the difference inthe specific gravity of the sample before and after removal of thedisulphidc by the addition of alcoholic potassium hydroxide. Afterremoval of carbon disulphide as xanthate, the proportions ofbenzene, toluene, and paraffin in the residue are found from thediflerences between the observed and recorded specific gravities asgiven in a table.70 The method is also applicable t o benzol firstrunnings, even when containing a high proportion of carbon disul-phide, but i t is advisable to dilute the liquid with a definite amountof purified benzene before the distillation.71 Another method ofestimating benzene and toluene in commercial mixtures is t o sepa-rate the constituents in pairs, benzene-toluene and toluene-xylene,by frzctional distillation, and t o obtain the composition of eachmixture from the boiling points.Curves showing the relationshipbetween the boiling points have been constructed from the respec-tive vapour pressures and checked by the results obtained withnixtiires of the pure substances.7211 new method of estimating phloroglucinol and resorcinol isbased on the fact that they yield insoluble precipitates whentreated, in hydrochloric acid solution, with f urf uraldehyde.Whenobtained under constant conditions as to acidity, volume of liquid,etc., the precipitates stand in definite relationship t o the quanti-ties ol these phenols, and may be washed, dried a t 100-105°, andweighed. Phenol, quinol, toluquinol, and catechol do not forminsoluble precipitates with the aldehyde, but cresols, xylenols, pyro-gallol, orcinol, and diresorcinol are precipitated by it .73Attention has been directed to sources of error in estimatingdicyanodiamide in calcium cyanamide by Caro's method.74 Precipi-tation of a small quantity of dicyanodiamide a t the same time as6 9 F. Robinson, J. Xoc. Dyers, 1916, 32, 81 ; A., ii, 366.7 0 P. E. Spielmann and E. G. Wheeler, J. SOC. Chew. Ind., 1916, 35, 396;71 P.E. Spielinann and F. B. Jones, i b i d . , 913 ; A . , ii, 583.72 A. Eclwa.rds, ibid., 587 ; A., ii, 462.i 3 E. VotoEek and R. Potmexil, Ber., 1916, 49, 1186; A . , ii, 512.74 N. Caro, Zeitsch. angew. Chem., 1910,23, 2407; A., 1911, ii, 162.A . , ii, 348178 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.the silver cyanamide cannot be prevented even by the addition ofammcinia in excess, whilst slight decomposition of the dicyaao-diamide takes place when the filtrate from the silver cyanamide isboiled with ptassium hydroxide. Hence the results given by themethod are too low, and the larger the proportion of dicyandiamidethe greatler will be the error.75 It is preferable to precipitate thecyanamide with neutral silver nitrate, and t o add the ammoniaafterwards, and to estimate the dicyandiamide by precipitating itsimultaneously with the cyanamide by means of silver nitrate and2 per cent. potassium hydroxide solution, and estimating the nitro-gen in the precipitate by Kjeldahl's method.76A useful paper dealing with the analysis of carbohydrates bymeans of enzymes and special yeasts was published early in theye2r.77 This gave a critical review of the various methods whichhave been proposed for the biological estimation of raffinose,maltose, and other sugars and starch, and described the most satis-factory means of obtaining active preparations of the enzymesfrom the specific micro-organisms.Incidentally, it is shown byexperiments t h a t reducing sugars are not actually precipitated bybasic lead acetate, b u t that laevulose, for example, is either destroyedby the lead salt or it is transformed into another sugar with adifferent specific rotatory power and a smaller reducing power.Hence in sugar analysis any considerable excess of basic leadacetate should be avoided, and such excess should be removed beforepolarisation.No loss of lzvulose occurs, however, unless the excess of lead isallowed to act on the sugar for some time before being precipi-tated.78I n estimating reducing sugars in the presence of a large excessof sucrose it is essential t o take1 into account such factors as thetemperature and the duration of heating, and to vary the amountsof sugar taken f o r the analysis according to the temperature andthe proportion of invert-sugar present .78aI n the case of products containing much invert-sugar, the cuprousoxide method of estimation is preferable, whilst for those poor ininvert-sugar the thiosulphate method gives the best results.The most suitable temperature for estimating the reducing powerof inverbsugar in presence of sucrose is 6 5 O , since a t t h a t tempera-ture the reducing capacity of sucrose is negligible, whereas a t higher'5 G.Hager and J. Kern, Zeitsch. angew. Ckem., 1916, 29, 309 ; A . ,76 E. Truniger, Schweiz. Ver. anal. Chem., May 26 and 27, 1916.7 7 W. A. Davis, J. Soc. Chew%. Ind., 1916, 35, 201 ; A . , ii, 202.7 8 W. A. Davis, J. Agric. Sci., 191 6, 8, 7.78a L. Maquenne, Compt. rend., 1916, 162, 207 : A., ii, 156.ii, 587ANALYTICAL CHEMISTRY.179temperatures, particularly from 90° to looo, there is a pronouncediiicrease in the reductioii.79Industrial sugars contain reducing sugars other than invert-sugar,slid these resemble sucrose in showing a greater cupric reducingpower a t boiling point (103-104°) than a t 6 5 O ; whereas in tliecase of invert-sugar there is only a slight increase in the reducingpower on raising the tenipe'rature from 6 5 O to boiling point. Henceby estimating the difference in the reduction caused by the indus-trial sugar at the t,wo temperatures, in comparison with that causedby sucrose, i t is possible to form an approximate conclusion as t othe amount of these unknown sugars in the product.80An experimental investigation of the method of estimatingpentose; and pentosans by means of Fehling's solution 81 has shownthat the modification devised by L.Eynon and J. H. Lane82 istrustworthy, but i t is suggested that by making two alterations inthe procedure, sources of error would be obviated: (1) by using alarger proportion of the distillate with a correspondingly greateramount of Fehling's solution, a greater weight of copper oxide isobtained, and errors due to multiplication are reduced; (2) byimmersing the flask in a boiling-water bath, instead of boiling thesolution, the reduction of Fehling's solution in the " blank " testis diminished by more than three-quarters. The large, spontaneousreduction which occurs when the Fehling's solution containing thesalt derived from the neutralisation of the acid is boiled alone, isprobably due t o the higher boiling point of the mixture.Thereduction due to the furfuraldehyde is not affected by the presenceof the salt, but the amount of copper oxide to be deducted fromthe t'otal obtained when tlie solution is boiled is inconveniently1arge.mMost of the contributions t o the analysis of oils and fats havedealt with the physical and chemical characters of less known pro-ducts, and there have been few additions to the known methods ofexamination.It has long been known that a definite relationship exists betweensome of the constants of fats, and a formula has now been estab-lished which includes the refractive index, ?a, specific gravity, d,saponiScation value, V , and iodine value, Z :-100 n2t - x ~n 2 t -I- 2 d:33-07 + 0-00075I- 0-01375V+ O-CO2 (t - 15).In the presence of hydrosy-acids the first figure of the equation7 9 L.Maquenne, Gompt. rend., 1916,162, 145 ; A . , ii, 156.8 o Ibid., 277 ; A . , ii, 202.81 J. T. Flohill, Chena. 'Ct'eekblad, 1910, 7 , 1057.82 Analyst, 1912, 37, 41 ; A . , 1912, ii, 305.83 J. L. Baker and H. F. E. Hulton, ibid., 1916, 41, 294; A , . ii, 651180 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.is lower. The relationship also holds good for hydrogenatedA method for estimating the degree of rancidity of oils and fatsis based on the proportion of aldehydes, etc., liberated on distilla-tion in a current of ste’am under constant conditions, and measuredby titration with permanganate solution.The difference betweenthe amount of permanganate required for the oxidation a,nd thatconsumed in a blank test gives the oxidisability value, which maybe defined as the mg. of oxygen required t o oxidise the volatileorganic compounds separated under constant conditions from thef at.85Speaking generally, the oxidisability value of sound, fresh fatsvaries from about 3 t o 10, whilst rancid fats may show values upto 50 or higher. There is no definite relationship between the acidvalue and the oxidisability value.The bromine thermal value 86 affords a rapid method of estimat-ing the iodine value of most unoxidised oils. I n a recent modifica-tion of the method 87 the heat of bromination is expressed in caloriesper gram of oil, and the relationship of the values t o the iodinevalues of a number of typical oils is given.Any pronounced differ-ences between the calculated and determine’d iodine values must beattributed t o there being a greater ainount of substitution by thebromine than by the iodine.Estimation of the stearic acid in fatty acids by crystallisationfrom a saturated alcoholic solution of pure stearic acid 88 sometimesgives abnormal results. These have been found to be due to super-saturation, as a result of insufficient stearic acid in the solution,and the error due t o this cause may be prevented by adding aknown quantity of pure stearic acid in excess of that requiredto saturate the alcohol, and deducting the weight of the, depositobtained in a “blank” experiment from that recovered in theactual estimation.89A rapid method of estimating glycerol as sodium glyceroxide inoils gives results in agreement with the amount of glycerol calcu-lated froin the ester value, provided that any free fatty acids in theoil are removed before the treatment with sodium ethoxide solutionand light petroleum.900iis.8481 H.J. Backer, Chem. Weekblad, 1916, 35, 954 ; A . , ii, 543.*j G. Issoglio, Ann. Chirn. Applicata, 1916, 6, 1 ; A., ii, 401.86 0. Hehner and C. A. Mitchell, Analyst, 1895, 20, 118.8‘ J. W. Marden, J . I n d . Eng. Chem, 1916, 8, 121.88 0. Hehner and C. A. Mitchell, Analyst, 1896, 21, 318.E. B. Holland, J. C. Reed, and J. P. Buckley, jun., J . Agric. Rescrirch,H. Bull, Chem.Zeit., 1916, 40, 690 ; A , , ii, 584.1916, 6, 101ANALYTICAL CHEMISTRY. 181The digitonin iiiethod of separating cholesterol and phytosterolis more trustworthy than the older method, and has the advantagethat i t can in most cases be applied, a t all events as a qualitativetest, t o the fats themselves. For a quantitative estimation it ispreferable t o apply the reagent t o the separated fatty acids. Evenfor the estimation of cholesterol in blood, precipitation with digi-tonin is more accurate than the colorimetric method.91A new method of separating and estimating cholesterol andisocliolesterol has been based on the precipitation of the formerby means of oxalic acid, and of the latter in the filtrate as abenzoate.92A method of estimating alkaloids as hydrochlorides is more simplethan many of the standard methods, while yielding equally accurateresults.The alkaloid is extracted with ether, hydrogen chloridepassed through the ethereal solution, the solvent evaporated, andthe residual hydrochloride weighed; or the residue may be dis-solvd in water, and the solution titrated with standard alkali solu-tion. With suitable modifications for the extraction, the methodgives trustworthy results in the estimation of alkaloids in suchproducts as conium seeds, colchicum root, and tobacco.93A critical examination of the methods of estimating nicotine intobacco94 has shown that precipitation of the alkaloid by means ofsilicotungstic acid 95 is trustworthy, and a simpler modification hasbeen suggested.Pyridine, which causes the results to be too high,is best separated from the nicotine by distillation with stelam froman acetic acid solution, which retains the nicotine. Accurateresults may be obtained by the methods of Kissling96 and ofKoenig,97 but the methods of Keller98 and of T6th99 are not sotrustworthy, and Ulex’s 1 method is untrustworthy. Thoms’ 2method of precipitation by a potassium-bismuth iodide reagent hasthe drawback that substances other than nicotine are also liableto be precipitated.J. H. Mueller, J. Biol. Chem., 1916, 25, 549 ; A., ii, 541.92 A. Madinaveitia and A. Gonzhlez, Anal. Pis. Quim., 1916, 14, 398 ;93 G. D. Beal and St. E. Brady, J. Ind. Eng. Chem., 1916, 8, 4 8 ; A.,94 H. B. Rasmussen, Zeitsch. anal.Chem., 1916, 55, 81 ; A., ii, 359.95 G. Bertrand and M. Javillier, Bull. Soc. chim., 1909, [iv], 5, 241 ;96 R. Kissling, Zeitsch. anal. Chem , 1882, 21, 64; A., 1882, 1005.97 W. Koenig, Chern. Zeit., 1911, 35, 521, 1047; A., 1911, ii, 670, 1143.9 * C. C. Keller, Ber. Deutsch. pharm. Ges., 1898, 8, 145; A., 1899, ii, 193.9 9 J. T6th, Chem. Zeit., 1911, 35, 146; A., 1911, ii, 334.2 H. Thoms, Ber. Deutsch., pharm. Ges., 1900, 10, 19; A., 1900, ii, 428.z4., ii, 585.ii, 356.A . , 1909, ii, 450 ; Ann. Chim. anal., 1911, 16, 251 ; A., 1911, ii, 827.H. Ulex, ibid., 121 ; A., 1911, ii, 33182 ANNUAL REPORTS ON THE PROGRESS OF CHEMTS'fKY.I i l o r g a n i c A tiulysis.&uaJitatiue.-In the ordinary method of separating silver frommercury by treating the mixed chlorides with ammonia, the blackprecipitate, which consists of a mixture of metallic mercury withthe mercuric compound, retains a small amount of the silver inthe form of an amalgam.To obviate this, the mercury in theprecipitate should be converted into the mercuric state by oxida-tion with bromine or sodium hypochlorite and nitric acid.3 Anammoniacal solution of ammonium perchlorate may be used as asensitive reagent for cadmium, which it precipitates as a white,crystalline, double perchlorate, [Cd(CIO,),NH,]. I n applying thetest to the mixed sulphides of copper, cadmifim, and bismuth, theprecipitate is dissolved in nitric acid and an excess of ammoniasolution added. The bismuth hydroxide is separated by filtration,and the blue filtrate is treated with the reagent, which precipitatesthe cadmium even in the presence of five times its quantity ofcopper.*For the detection of arsenic, the silver nitrate test is verysensitive, a brown coloration being obtained with a solution con-taining as little as 1 part in 150,000, whilst a distinct precipitateis given by solutions containing 1 part in 60,000.Even when thetest is applied in the presence of ammonium nitrate, as formed inordinary routine analysis, it is still capable of detecting 1 part ofarsenic in 15,000.5The dimethylglyoxime test for nickel in the presence of largeamounts of cobalt may be rendered much more sensitive by addingalkali cyanide to the solution of the mixed nickel and cobalt saltsuntil the resulting precipitate begins t o dissolve, after which theliquid is heated and rotated until its colour changes.It is thendiluted with hot water and treated with the reagent, and silvernitrate solution is added drop by drop to precipitate the cyanideas silver argenticyanide and promote the decomposition of thenickelocyanide ions. Under these conditions, the test is capableof detecting 0.02 mg. of nickel, or a tenth of that quantity if themixture is allowed to remain for twenty-four hours.6A very delicate test for manganese in the presence of largequantities of iron has been based on its oxidation to the manganiccondition by means of nitrous acid and the subsequent formationof manganic oxalate, which is of a bright red colour.The test3 N. von Zweigborgk, Zeitsch. anorg. Chem., 1916, 93, 3-30 ; A . , ii, 344.5 L. J. CurtmanandP. Daschnvsky, J. Amer. Chem. SOC., 1916, 38, 1280 ;R. Salvadori, Ann. Ghim. Applicata, 1916, 5, 2 5 ; A., ii, 271.A . , ii, 491.A. R. Middleton and H. L. Miller, ibid., 1705 ; A . , ii, 580ANALYTICAL CHEMISTRY. 183may be conveniently applied by adding an excess of neutral sodiumnitrite solution to a neutral solution of the manganous salt andtreating the mixture with oxalic acid.7Tungsten in the proportion of 2 per cent. or less in mineralsmay be detected by the blue coloration which it gives on reduc-tion with metallic tin in hydrochloric acid solution. Columbiumgives a similar blue coloration, but may be distinguished by thefact that the colour disappears on dilution, whilst the colorationgiven by vanadium may also be produced by reduction withtartaric acid, which is not the case with tungsten.Titanium givesa violet coloration under the same conditions.8 With suitableprecautions as t o the concentration of the acid, the test' is capableof detecting 1 mg. of tungstic acid.9The deposit of selenium, obtained by reducing selenious acid andselenites with zinc and sulphuric acid in Marsh's apparatus, is ofa red colour quite distinct from the colour of an arsenic deposit.To distinguish between arsenic and selenium in larger quantities,the hot solution may be treated with hydrogen sulphide. Theprecipitated sulphur, containing arsenic and selenium sulphides ifpresent, will be brown in the presence of selenium, whilst its colouris not affected by arsenic sulphide.By heating the dry precipitatein a tube, the free sulphur is volatilised, leaving the seleniumsulphide as a black mass.A test for nitrates in the presence of organic matter dependson the fact that an ethereal extract of a solution of salicylic acidin concentrated sulphuric acid gives no red coloration with ferricchloride solution in the absence of nitrates. If nitrosalicylic acidis present, an orange coloration is produced on adding ammoniato the ethereal extract, and a red coloration in the aqueous layeron the addition of dilute ferric chloride solution.11Quantitative.-The advantages of potassium dichromate as astandard for volumetric methods are that the salt can be readilypurified and that its solutions are stable, but it has been statedby Wagner and others that it gives too high results on titration.This discrepancy has now been found to be due, in part a t least, tothe older atomic weights being inaccurate, for comparative estima-tions have shown that the oxidation follows a course similar tothat' effected by potassium permanganate.The slightly high results(0.13 per cent.) are attributable, not to any property inherent inthe dichromate, such as its promoting catalytic oxidation, but t oW. Prandtl, Ber., 1916, 49, 1613 ; A . , ii, 621.M. L. Hartman, Chem. News, 1916, 114, 2 7 ; A., ii, 494.M. L. Hartman, ibid., 45; A., ii, 495.lo J. Bieunier, Compt. rend., 1916, 165, 332 ; -4., ii, 6.41.l1 A.Tingle, J . Xoc. Chem. Ind., 1916, 35, 77 ; A., ii, 195184 ANNIJAL REPORTS ON THE PROGRESS OF CHERIIS'I'RY.differences in the methods of titration. The best method ofstandardisation is gravimetric estimation with oxalic acid. It isnot possible t o use dichroniate as a standard acid in alkalimetry,since the acidity of the solution increases with the concentra-tion.12A convenient method of estimating certain bivalent metals isto titrate their phosphates with standard acid.13 In the case ofmagnesium, i t is advisable to use 50 per cent. alcohol in place ofthe ammoniacal solution for washing the precipitated magnesiumammonium phosphate. Zinc ammonium phosphate and cadmiumammonium phosphate may also be directly titrated in the sameway, cochineal being used as indicator, but the method gives un-satisfactory results with manganese.The precipitated cobaltammonium phosphate varies in properties with the mode of pre-paration. The precipitate obtained by Clarke's method 14 is suit-able for direct titration with acid, but a double precipitation isadvisable t o remove traces of nickel. The nickel in the filtratemay be accurately estimated by titration with cyanide solution.It is possible to utilise the reducing action of titanium chloridein several ways in volumetric analysis. Thus, ferric chloride andchromates may be accurately titrated in hot hydrochloric acid solu-tion with niethylene-blue as indicator, whilst cupric salts may beestimated in the same way with safranine or induline as indicator.F o r the estimation of titanium in ores, the sample is fused withalkali, dissolved in hydrochloric acid, and the solution reducedwith zinc and treated with standard ferric chloride solution, theexcess of which is titrated with titanium trichloride in an atmo-sphere of carbon dioxide.l5From the results of a study of the oxidations effected bypotassium permanganate in alkaline solution, it appears that theproducts may vary with the degree of alkalinity.For example,in a slightly alkaline solution, nickel is oxidised to an oxide,NiloOll, whilst in strongly alkaline solution an oxide approximatingin composition t o Ni,O, is produced. Oxidation of cobalt instrongly alkaline solution yields an oxide containing slightly moreoxygen than is required by the formula Co,O,.I n the titrationof arsenious acid in strongly alkaline solution, the permanganateis reduced t o the manganic condition when potassium sulphate orother electrolyte is present t o cause precipitation of colloidall2 G. Bruhns, J. pr. C'hem., 1916, [ii], 93, 73, 312 ; A . , ii, 337, 581.l3 W. R. Schoeller and A. R. Powell, Analyst, 1916, 41, 124; A . , ii, 846.Chem. News, 1883, 48, 262.A. Monnier, Ann. Chim. anal., 1916, 21, 109; A . , ii, 444AXALYTICAL CHEMISTRY. 185manganese hydroxide. Should the colloidal hydroxide remain insuspension, however, the reduction of the manganese continues tothe manganous condition. Thallous and cerous salts are quanti-tatively oxidised to thallic and ceric salts, and selenious andtellurous acids to selenic and telluric acids respectively.I n thecase of lead, however, the oxidation is incomplete.16Several useful applications of iodometric methods have beenbrought forward during the year. A differential method of esti-mating periodates, iodates, bromates, and chlorates has been basedon the different conditions under which they react with iodinesolutions. For example, periodates are quantitatively decomposedinto iodates and free iodine when treated with a solution of iodinein a saturated solution of boric acid containing borax to reducethe acidity. Both iodates and periodates react with a solution ofiodine in N/4-acetic acid, and the liberated iodine may be titrated.A dilute (AT/2) hydrochloric acid solution of iodine reacts withbromates, whilst a solution of 6X-strength reacts with chlorates,leaving perchlorates unaffected.l7 The presence of hypochloritesdoes not affect the accuracy of the results obtained in the estima-tion of chloric acid and chlorates by this method.lsIt has long been recognised that thiosulphates and sulphitesare formed when tetrathionate solutions are treated with hot con-centrated alkali hydroxides, and it has now been shown that thereaction also takes place fairly rapidly in cold and dilute solu-tions.19 Hence, in titrating solutions containing tetrathionate, itis essential not t o render the liquid alkaline.Should this betemporarily unavoidable, as, for instance, in the case of a solutionto which iodine and excess of thiosulphate have been added, it isbest t o use the smallest possible amount of sodium carbonate andto keep the temperature low.It is also advisable to saturate theliquid with carbon dioxide.I n the iodometric estimation of gold by Peterson’s method,20the results are rendered more accurate by treating the compound,AuCl,,HCl, with potassium iodide and iodate and titrating theliberated iodine. The gold salt is first converted into gold chlorideand free hydrochloric acid, and the latter reacts with the iodate,forming free iodine and potassium chloride. Hence, in analysinggold chloride, AuC13,H,0, by this method, the difference in theresults obtained on titrating the iodine liberated from potassiuml6 B. Brauner, Zeitsch, anal.Chem., 1916, 55, 225 ; A . , ii, 437.l7 0. L. Barnebey, J. Amer. Chem. SOC., 1916, 38, 330 ; rl., ii, 261.lR R. L. Taylor, J. Xoc. Dyers, 1916, 32, 6s; A . , ii, 193.l D R. M. Chaplin, J. Amer. Chem. Soc., 1916, 38, 625 ; A . , ii, 261.2 o H. Peterson, Zeitsch. anorg. Chem., 1898, 19, 59; rZ., 1899, ii, 250186 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.iodide and from a mixture of potassium iodide and iodate affordsa measure of free hydrochloric acid if present.21I n the absence of reducing compounds, such as ferrous, cuprous,and antimonous salts, or of precipitated metals, tin in thestannous condition may be titrated with potassium iodate solution.The reagent is run into the strongly acid solution of the stannoussalt, to which has been added a little chloroform, and the stopperedbottle is shaken after each addition until the chloroform, whichhas absorbed liberated iodine, is decolorised.Stannic solutionsare conveniently reduced with metallic nickel. With suitablemodifications, the method is applicable to the analysis of solders,bronzes, etc.22A rapid method of estimating mercury is based on the reductionof mercuric salts by means of formaldehyde in the presence ofsodium hydroxide or potassium hydroxide. The reduced metal isthen quantitatively converted into mercuric iodide by treatmentwith standard iodine solution, the excess of which is titrated i t 1the usual way.23Cobalt salts may be readily oxidised by means of hydrogenperoxide or sodium perborate, preferably the latter, and on thisreaction has been based a method of estimating cobalt.Afterremoval of the excess of the oxidising agent, the solution isacidified and treated with an excess of potassium iodide, and theliberated iodine is titrated with thiosulphate solution which hasbeen standardised on a solution of a pure cobalt salt. The methodis applicable in the presence of nickel and alkalis, but the solutionmust not contain iron, manganese, or any substances capable ofliberating iodine from potassium i0dide.~4A volumetric method of estimating thiosulphates in the presenceof sulphites depends on the fact that mercuric chloride reacts withsulphites t o form soluble double salts, whilst i t decomposes thio-sulphates, with the liberation of free sulphuric acid, the titrationof which affords a measure of the thiosulphate.The precipitationof mercuric oxide by the alkali during the titration is preventedby the addition of ammonium chloride.25A new method of estimating fluorine is based on its precipita-tion as thorium fluoride from an acid solution and ignition of theprecipitate, which leaves a residue of thorium oxide. The chief21 L. Vanino aid F. Hartwagner, Zeitsch. anal. Chewi., 1916, 55, 377 ; A . ,22 G. S. Jamieson, J . Jnd. Eng. Ghem., 1916, 8, 500; A . , ii, 451.23 G. Adanti, Boll. chim. fcirm., 1916, 55, 563 ; A . , ii, 579.24 W. D. Engle and R. G . Gustctvson, J . I n d . Xnq. Chew., 1916, 8, 901 ;2j A. Sander, Zeitsch. anal. Chem., 1916, 55, 340 ; A,, ii, 536.ii, 582.A., ii, 649ANALYTICAL CHEMISTRY.187precaution necessary is not to add too large an excess of thoriumnitrate, which tends t o redissolve the precipitate. Silicofluoridesmay be precipitated directly in the same way, whilst tantalo-fluorides should be boiled with sodium carbonate and filtered fromthe tantalic acid before precipitation of the fluorine.26The low results sometimes obtained in the estimation of silicain silicates are partly due t o the fact that silica is somewhatsoluble in solutions of sodium chloride. To obviate this, i t isnecessary to use as small a proportion as possible of sodiumcarbonate for the fusion of the silicate. Silica is also sparinglysoluble in hydrochloric acid, and hence one evaporation does noteffect a complete separation.Acid of specific gravity 1.1 has lesssolvent action on silica than stronger or weaker acid. The firstevaporation should be carried out as rapidly as possible, whilstthe final dehydration should not occupy more than two hours ata temperature not exceeding l l O O . 2 7I n the estimation of nitrogen by Kjeldahl’s method, the use ofmercury to accelerate the conversion causes the results to be toolow in the case of certain compounds, such as caffeine, uric acid,and pure ammonium sulphate. By using copper foil in place ofmercury, the correct results are obtained with ammonium sulphate,and the loss of nitrogen in the case of the other compounds isgreatly reduced.28The method of distilling the ammonia by the introduction of acurrent of cold purified air 2s gives good results in the case of manycompounds, but in other cases the yield of nitrogen is from 0.3 to1.97 per cent.lower than that obtained by the ordinary methodof distillation. For the analysis of substances of unknown com-position, the method must therefore be regarded as untrust-worthy .30The use of magnesia, which is suggested as the most suitablealkali for .the decomposition of the ammonium sulphate in thisprocess,31 is open t o the objection that part of the ammonia maybe retained in the distillation flask as ammonium magnesiumphosphate.As commonly used, the method of estimating arsenic by reduc-tion with ferrous salts and distillation as trichloride, frequentlygives low results. To render the method trustworthy, it is essential26 F.Pisani, Cornpt. rend., 1916, 162, 791 ; A . , ii, 393.27 V. Lenher and E. Truog, J . Amer. ChenL. Xoc., 1916, 38, 1050; A . ,28 0. Nolte, Zeitsch. anal. Chem., 1916, 55, 186 ; A . , ii, 341.2 9 P. A. Kober, J . Anzer. Chem. Soc., 1908, 30, 1131 ; .4., 1908, ii, 576.30 K. G. Falk and K. Sugiura, ibid., 1916, 38, 916 ; A . , ii, 341.s1 R. S. Davisson, E. R. Allen, and B. M. Stubblefield, ibid., 896; A . , ii, 643.i i , 396188 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.that only a small proportion of arsenic should be present, andcuprous chloride should be used in place of ferrous chloride as thereducing agent. The arsenic trichloride may then be distilled ina current of hydrogen chloride, and the distillate renderedalkaline, and, after the addition of sodium hydrogen carbonatein excess, titrated with standard iodine solution.32A new method of estimating cadmium depends on its precipita-tion as dipyridine cadmium chloride, CdC1,,2C5H5N, which is con-verted into the monopyridine compound when heated at 115O to120O. The latter is stable, and can be dried at temperatures upto 140° without decomposition. The corresponding cupric pyridinecompound is fairly readily soluble in pyridine solutions, and thisdifference from the cadmium compound may be utilised in thequalitative separation of the two metals, although the quantitativeresults thus obtained are only approximately correct4.33A modification of Volhard’s method is especially suited for theestimation of copper and iron in ores, since it enables both metalsto be estimated in the same solution. The copper is precipitatedas cuprous thiocyanate and the iron is titrated with potassiumpermanganate solution.I n preparing the ores, the sample isdecomposed with nitric and hydrochloric acids in the usual way,these acids being subsequently expelled by heating the liquid withsulphuric acid, after separation of any silver chloride.34It has been accepted that in estimating carbon in steel, higherresults are obtained by increasing the temperature of the combus-tioii in oxygen, although the conclusion does not appear t o havebeen based on trustworthy experiments. A recent investigationin which the temperature was maintained a t 1500O has shown thatthe amounts of carbon thus obtained agreed within 0.015 per cent.with those given by more simple methods of direct combustion.3sBaudisch’s I‘ cupferron ” reagent (ammonium salt of nitroso-phenylhydroxylamine) has been found applicable to the separationof metals other than those previously described.Thorium is quan-titatively precipitated by i t from an acetic acid solution, but theresults are not claimed to be as trustworthy as those given by theoxalic acid method.36 Thorium differs in this respect from zircon-ium, which requires the presence of sulphuric acid for its quantita-tive precipitation by the reagent.3732 R. C. Roark and C. C. MacDonnell, J . I n d . Eng. Chem., 1916, $, 327 ;33 S. Kragen, Monatsh., 1916, 37, 391 ; A . , ii, 647.34 G.Edgar, J . Amer. Chem. Soc., 1916, 38, 884 ; A . , ii, 346.35 J. R. Cain and H. E. Cleaves, J . I n d . Eng. Chem., 1916, 8, 321 ; L4.,36 W. M. Thornton, jun., Chem. News, 1916, 114, 1 3 ; A . , ii, 495.37 W. M. Thornton, jun., and E. M. Hayden, jun., Amer. -7. S c i . , 1914, [iv],-4., ii, 342.ii, 343.38, 137; A , , 1914, ii, 779ANALYTICAL CHEMISTRY. 189‘‘ Cupf erron ” also precipitates vanadium quantitatively from acidsolutions of ammonium metavanadate, and on igniting the precipi-tate a residue of the oxide, V,O,, is left. It is essential that thesolution should not contain more than 1 per cent. of hydrochloricor sulphuric acid.33With suitable modifications the method is also applicable to theseparation of vanadium from phosphoric and arsenic acids, andfrom uranium, but in presence of the latter the results for vanadiumare a little high.The uranium may be precipitated from thefiltrate by ammonia (after addition of ammonium chloride), andthe results thus obtained are accurate.39andfound t o yield accurate results in the analysis of ferro-tungsten.With slight modifications it is also suitable for tungsten concen-trates containing up t o 10 per cent. of tin. Any traces of tin leftin the final precipitate of tungstic acid are expelled by volatilisa-tion with ammonium chloride.41Potassium in the presence of large amounts of sodium may bemost accurately estimated by precipitating i t with cobalt nitrite,*:dissolving the precipitate in hydrochloric acid, and making thefinal estimation by the perchlorate method, thus eliminating sub-stances wliich interfere with the accuracy of the latter.43 The com-bined method is not applicable in the presence of ammonium salts,whilst phosphates of iron or aluminium, if present, should be keptin solution by the addition of sodium citrate prior to the precipita-tion of the potassium with cobalt nitrih44Fieber’s method of estimating tungsten has been testedElectrochemical ,4nalysis.A useful method for titrating vanadates with ferrous sulphatesolution has been devised, in which the change in the electromotiveforce of a galvanic battery is used to indicate the end-point of thereaction.The apparatus comprises a resistance box of two coilsand sliding contacts arranged so as t o vary the E.ik1.F.of the drycells. The solution of the vanadate is treated with dilute sulphuricacid, and the ferrous sulphate solution run in until the galvano-meter needle, which becomes practically steady towards the end ofthe titration, makes a sudden move, indicating the addition ol an38 W. A. Turner, Amer. J. Sci., 1916, [iv], 41, 339; A., ii, 347.39 W. A. Turner, ibid., 1916, [iv], 42, 109 ; A., ii, 540.4 0 R. Fieber, Chem. Zeit., 1912, 36, 334; A., 1912, ii, 495.‘l E. Dittler and A. von Graffenried, ibid., 1916, 4.0, 681 ; A , , ii, 552.42 E. Mitscherlich and H. Fischer, Landw. Versuchs-Stat., 1912, 78, 74 ;43 R. G. Thinand A. C. Cumming, T., 1915,107, 361 ; A., 1915, ii, 281.44 A. H. Bennett, Analyst, 1916, 41, 165; d., ii, 448.A., 1912, ii, 996190 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.excess of tlie reagent.The correct figure is then found by titratingthe liquid with standard dichromate solution until the needlereturns t o the place where i t was before its sudden move. Themethod gives results in close agreement with those obtained by tlieusual standard methods.45By the use of suitable modifications the method is also applicableto the estimation of chromium and vanadium in steel.46It has already been shown 47 that copper and tin may be precipi-tated by an electric current from their solution in nitro-hydrofluoricacid, leaving lead and antimony in solution, and later experimentshave shown that the same method may be used for separating silverand mercury from tin, antimony, tungsten, and molybdenum.Thesilver deposits usually contain no appreciable quantity of platinum,but the mercury deposited on the1 anode usually contains a con-siderable proportion of platinum, the amount depending on suchfactors as the quantities of tin and hydrofluoric acid, and the dura-tion of the electrolysis. The necessary correction may be obtainedby treating the deposit with nitric acid and estimating the undis-solved platinum.@It is commonly accepted t h a t i t is not practicable to obtain gooddeposits of metals from solutions of their chlorides owing to the sub-sequent oxidation caused by the liberated chlorine, but recentwork has shown that under suitable conditions this action niay beprevented.49 To obtain satisfactory electrolytic deposits of anti-mony, bismuth, copper, lead, tin, and cadmium from solutions ofthe metals in hydrochloric acid, i t is necessary t o add a suitablereducing agent, such as hydroxylamine hydrochloride, and t o seethat no nitric acid or oxides of nitrogen are present.Silver may also be estimated in the same way by electrolysingsolutions of silver chloride in ammonia.The deposits are purerthan those obtained in the electrolysis of cyanide solutions, and themethod gives trustworthy results.50An electrolytic method of estimating zinc gives good results,provided that certain conditions are maintained, such as the degreeof acidity of the electrolyte and the uniform density of the currenton the cathode.51 The zinc is converted into sulphate, and is de-posited from dilute sulphuric acid solution t o which has been added45 G.L. Kelley and J. B. Conant, J . Amer. Chem. Xoc., 1916, 38, 341 ;413 G. L. Kelley and J. B. Conant, ibid., 719 ; A., ii, 540.47 L .W. McCay, ibid., 1914, 36, 2375 ; A . , 1914, ii, 856.48 L. W. McCay and N. H. Furman, ibid., 1916, 38, 640: A . , ii, 273.E. P. Schoch, D. J, Brown, and T. E.Phipps, ibid., 1660 ; A , , ii, 578.6 0 E. P. Schoch and F. M. Crawford, ibid., 1682 ; A . , ii, 576.61 F. Chancel, Bull. SOC. chim., 1916, 19, 59 ; A., ii, 198.A., ii, 274ANALYTICAL CHEMISTRY. 191a quantity of sodium formate corresponding with the amount ofmetal. The cathode consists of sheet platinum and the anode ofplatinum wire.52 The method is also applicable t o brass afterremoval of iron by precipitation.lVa t e r A nalysis.For the estimation of dissolved oxygen in water, Winkler’s modi-fied method 53 has stood the test of critical examination from severalquarters, and may now be accepted as generally trustworthy.Themodifications suggested during the past year have not claimed t oyield more accurate results, but have had for their object thesimplification of working details. Such modifications, for example,are the use of potassium iodide in solid form, and the titration ofthe liquid in the original bottles.54 The presence of nitrites ororganic matter in the water interferes with the estimation. Onemethod of eliminating this influence is t o treat the water. withdilute sulpliuric acid and a slight excess of calcium hypochloritesolution prior t o the estimation of the oxygen by the manganouschloride method.A t the same time a blank test is made to ascer-tain the effect of the1 excess of chlorine on the results of the titra-tion, by treating an equal volume of the water with the same quan-tities of sulphuric acid and calcium hypochlorite, adding a trace ofpotassium iodide, and titrating the liberated iodine.55Another means of inhibiting the effect of nitrites and organicmatter is t o convert the manganous hydroxide into carbonate by theaddition of solid potassium hydrogen carbonate, t o collect the pre-cipitate, and to treat i t with potassium iodide and acid.56 If ferroussalts are present, they should be oxidised by means of permangan-ate, and the water should then bel acidified with phosphoric acidinstead of with hydrochloric acid.56 None of the suggested modi-fications is trustworthy when organic matter is present in largequantities, as in certain trade efnuents, and in such cases a gas-volumetric method of estimating the dissolved oxygen must beused.57A simple method of detecting methane in the gases separatedfrom water consists in first absorbing the oxygen in the usualmanner, and then treating the residue with isobutyl alcohol whichhas previously been saturated with air.Methane, if present, will52 See F. Chancel, Bull. Xoc. chim., 1913, Liv], 13, 74 ; A., 1913, ii, 236.53 L. W. Winkler, Zeitsch. anal. Chem., 1914, 53, 665; A., 1915, ii, 277.51 G.Bruhns, Chem. Zeit., 1915, 39, 845 ; -4., 1916, ii, 47 ; Chem. Zeit.,1916, Po, 45, 71 ; A., ii, 146.56 L. W. Winkler, Zeitsch. angew. Chem., 1916, 29, i, 44 ; A., ii, 194.56 G. Bruhns, Chem. Zeit., 1916, 40, 45, 71 ; A., ii, 146.b7 L. W. Winkler, Zeitsch. Nahr. Genussm., 1915, 29, 121 ; A., ii, 487192 ANNUAL REPORTS ON THE PROGRESS OF CHEMIISTHY.be absorbed by the reagent, and there will be a decrease in thevolume of the gas, whereas the volume of nitrogen will not beaffected.58Rapid methods of sterilising water are of paramount importancea t the present time, especially such as are applicable for the use ofarmies in the field. I n one of these methods a reagent containingactive chlorine is prepared by treating a solution of sodium hypo-chlorite with potassium permanganate, and the excess of chlorinein the water is subsequently destroyed by means of thiosulphate.59Any traces of thiosulphate left in the water may be detected andcolorimetrically estimated by means of silver nitrate, followed byjust sufficient ammonia to dissolve all silver chloride.60 A sensitivereagent for free chlorine in water which has been treated withalkaline hypochlorite is an acid solution of hexamethgltri-p-amino-triphenylmethane, which gives an immediate violet coloration withwater containing as little as 3 parts of free chlorine in 100,000,000.An advantage which this reagent has over the potassium iodide-starch reagent for free chlorine is that it is not readily affected bynitrites.61Of late years there has been a tendency to attach undue weightt o the results of oxygen absorption tests in the analysis of water,and in some cases when the conclusions to be drawn from thegeneral results have been doubtful the permanganate test has beenmade the decisive factor.Although there can be no question as t othe value of the test f o r following any variations in the purity ofa water supply, the uncertainty in the results due to such factorsas the temperature, duration of action, and the amount of excessof permanganate renders the method untrustworthy for a water ofunknown origin.62In any case, oxidation in alkaline solution must be regarded asliable to grave error from the fact that there is a tendency for thepernianganate to be reduced to rnanganate through the catalyticaction of the bulky precipitate of manganese dioxide.63 Hence, theresults obtained with an acid permanganate solution are more trust-worthy, notwithstanding the fact that the effect of chlorine in thewater is minimised by the use of an alkaline solution.Under anyconditions, however, it is preferable to remove the chlorine by meansof silver oxide before estimating the oxygen absorption. The influ-5 8 L. W. Winkler, Zeitsch. angew. Chem., 1916, 29, i, 218; A., ii, 448.59 H. Penau, J. Pharm. Chim., 1916, [vii], 13, 377.6 o J. Golse, ibid., 1916, 14, 8.61 G. A. Le Roy, Compt. rend., 1916, 163, 226; A., ii, 535.62 G. W. Heise and R. H . Aguilar, Philippine J. Sci., 1916, [A], 11, 37 ;63 J.H . Sachs, J. I n d . Eng. Chem., 1916, 8, 404; A., ii, 399.A., ii, 576AN A L P‘I’IC AL CH E hI I STK Y. 193eiice of tlie quantity of the reagent is most pronounced a t highertemperstures. A t 85O the absorption of oxygen increases with theamount of permanganate added, whereas a t 3 7 O variations in thequantities added have but little effect on the results. This tem-perature is therefore the best f o r accelerating the absorption whilsteliminating t o a considerable extent the quantity factor.64Recently proposed methods of estimating the hardness of watermust be regarded as, in the main, either modifications of Hehner’smethodG5 or of Clark’s soap test, in which an alkali salt of a purefatty acid takes the place of the soap solution. I n Wartha’smethod of estimating the alkalinity by titrating the boiling waterwith standard hydrochloric acid, the use of alizarin as indicatorgives sharper results than when methyl-orange is used, althoughwith the latter indicator there is no need t o boil the water.66 Ironsalts, if present, must be eliminated prior t o the titration.A convenient modification for estimating the total hardness isto evaporate the water with standard potassium carbonate solution,to extract the residue with 90 per cent. alcohol, and to titratetlie filtered extract with standard hydrochloric acid, with methyl-orange as indicator. This method is based on the facts thatcalcium and magnesium carbonates are practically insoluble inalcohol of that strength, whilst the double carbonate of magnesiumand potassium which may be formed is decomposed, with precipita-tion of the magnesium carbonate and solution of the potassiumcarbonate, which, unlike sodium carbonate, is fairly readily solublein strong alcohol.67Blacher’s method,68 in which pure potassium palmitate is usedf o r titrating the total hardness, has proved more generally applic-able than methods involving the use of salts of other fatty acids.It has the drawback that after the temporary hardness has beenestimated by titrstion with standard hydrochloric acid, it is noteasy to ascertain whether all the free carbon dioxide has beenexpelled. This is easily obviated by neutralising the free carbondioxide with N / 10-potassium hydroxide solution, with phenol-phthalein as indicator, prior t o the titration with potassium palmi-tate solution.69As a rapid means of checking the chemical estimation of thehardness, the electrical conductivity of the water may be deter-64 J. H. Sachs, loc. cit.L. ’IV. Winkler, Zeitsch. angew. Chenz., 1916, 29, i, 2 1 8 ; A . , ii, 448.67 S. A. Kay and S. H. Newlands, J . SOC. Chem. I d . , 1916, 35, 445 ; A . ,C. Blacher, P. Griinberg, and M. Kissa, Chenz. Zeit , 1913, 37, 5 6 ; A . ,65 0. Hehner, Analyst, 1883, 8, 77.ii, 344.1913, ii, 153.69 RI. Tilgner, ibid., 1916, 41), 675 ; A . , ii, 57’7.REP.-VOL. XIII. 194 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.mined by means of an apparatus such as the “dionic watertester.” XI There should be an approximate relationship betweeiithe two results, each 20 units of electrical conductivity correspond-ing with one degree of hardness.71For the separate estimation of calcium and magnesium, advan-tage may be taken of the fact t h a t whereas magnesium carbonateis soluble in an excess of ammonium carbonate solution, calciumcarbonate is almost insoluble therein, and may be estimated in theinsoluble residue by titration with standard acid. The differencebetween the result and the total hardness determined by the alcoholmethod (see above) will give the amount of magnesium.72C. AINSWORTH MITCHELL.7 0 L. Archbutt, Analyst, 1912, 37, 538.71 S. A. Kay and S. H. Nowlands, J . Soc. Chem. Id., 1916, 35, 445; il.,72 S. A. Kay and S. H. Kewlands, ibid., 447 ; A . . i i , 345.ii, 344

ISSN:0365-6217    

DOI:10.1039/AR9161300165

出版商:RSC    

年代:1916

数据来源: RSC

摘要:

PHYSIOLOGICAL CHEMISTRY.ARRHENIUS, as is well known, has applied the accurate methods ofphysical chemistry to the complex phenomena of immunity, and, ina book recently published in English,l he has brought together manyof the quantitative data which exist in the literature concerningthe reactions of the animal body to antigens. He has treatedthem, as well as the subject of enzymic action in general, fromthe mathematical point of view of the physical chemist. It isremarkable to realise how much of the available material provesitself amenable to such treatment. The writings of so great ateacher as Arrhenius cannot but have a very healthy influenceon biochemistry. I venture t o think, however, t h a t he takes tooone-sided a view with regard to the methods likely to lead toprogress in the immediate future.I n the preface to his book he writes : “The aversion shown b jbiochemists, who have in most cases a medical education, to exactmethods is very easily understood.They are not acquainted withsuch elementary notions as ‘ experimental errors,’ ‘ probableerrors,’ and so forth, which are necessary for drawing valid con-clusions from experiments.”Under t h a t criticism many biochemists of to-day, although byno means all, must-with certain self-defensive reservations-bowtheir heads. When, however, a science is largely in its descrip-tive stage, it would be a pity if any of its devotees allowed them-selves to be too much discouraged by particular deficiencies intheir education, especially when that education happens to beone which, a t any rate, makes them, better than anyone else, ableto appreciate the nature of the problems to be solved.We allknow t h a t to arrive a t the mathematical form is the ultimate goalof all real scientific knowledge; but a t a given moment in thehistory of a science, the acquirement of new qualitative knowledgemay be as important as the consolidation of other knowledge in amathematical form. I n biochemistry, which is just now busy i nLondon : G Bell & Sons, “ Quantitative Laws in Biological Chemistry.”1915.H 2 19196 ANNUAL REPORTS ON THE PROGltESS OF CHEMISTRY.collecting its data, the mere discovery of hitherto unknowii reac-tions in the animal body, with a supply of information concern-ing their locality and their significance to the body as a whole,may well lead to more profitable suggestions for work in theimmediate future than the proof that this or that reaction isstrictly unimolecular, or that the laws of equilibria apply withexactness in such cases as the binding of toxins by antitoxins.It is, I think, only honest to admit that whilst the quantitativemust in some sort enter into every investigation, to demand it ina particular form a t a given moment may savour of the doctrin-aire.Moreover, the methods of physical chemistry, when appliedto phenomena in which what we call “physical” does not com-pletely outweigh what we still call “chemical,” may, in arrivingby a short cut a t accuracy of statement, leave us ignorant of justthose details which are most necessary for the guidance of furtherresearch.To biology in its present position the chemical detailsare of great importance.We know, it is true, almost nothing about the pure chemistryof the materials and agents which play their part in the pheno-mena of immunity, or in such events as the coagulation of theblood. For this reason, it is perhaps scarcely fair to appraise theresearch done on them under headings avowedly chemical; butthe chemical knowledge is urgently desired, and it will come thesooner if the attention of chemists be directed to the problems.The Coagulation of t h e Blood.This subject received treatment from Professor Halliburton inthe Report for 1912,z but in order to make clear the significanceof the progress recently made, it seems necessary t o give a verybrief rBsum6 of the present position, even although it involvessome repetition of what has already appeared in these Reports.Any divergence in the views a t present held with regard tothe essential nature of the process (as distinct from its details)arises from a difference in the degree of importance ascribed t ocatalytic factors.The older view, based upon the researches ofBuchanan and Alexander Schmidt, was developed by Hammar-sten, and expanded a few years ago by Morawitz, Fuld,5. Mellanby, and others. It ascribes the initiation of coagulatioato the appearance in shed blood of catalytic agents which areeither absent from the circulating fluid (being held in the formedelements and the tissues) or are there balanced by anti-coagulativeagents.This view involves the belief that chemical change pre-cedes clot formation.2 See Ann. Repoyt, 1912, 236PHYSIOLOGICAL CHEMISTRY. 107From the other point of view, blood coagulation is consideredas due to a rearrangement in an unstable system of associatedcolloids, initiated perhaps by the addition of fresh colloids fromthe formed elements of the blood or from tissue cells. Thisrearrangement may involve some change in pre-existing looseassociations between colloid and colloid, but not necessarily achemical change in any strict sense. The process is thus essenti-ally physicochemical in nature. Although the fact is not alwaysrecognised, chiefly because there is so great a breach between theterminology of to-day and t h a t of thirty years ago, this view owesits parentage t o Woolridge.It was much later supported in thesomewhat involved publications of Nef, and has received recentsupport from America, especially in papers from the laboratoryof W. H. Howell. I: note that A. P. Mathews gives strong sup-port to the conception of blood coagulation as a purely physico-chemical process in his excellent ‘‘ Text-book of PhysiologicalChemistry,” which, although it was published last year, receivedno reference in my last Report.So brief a statement as the above may make the cleavage ofopinion seem sharper than it really is, but I think it fairly repre-sents two existing currents of thought about the matter.It is still generally agreed that the following factors are con-cerned in the phenomenon: first, the fibrinogen of the plasma,the essential material precursor of the clot; secondly, an agent,thrombin, which is essential for the conversion of fibrinogen intofibrin.This has long been looked upon as an enzyme, but Howellholds t h a t it should be removed from that category because, accord-ing to his observations, increasing amounts of thrombin giveincreasing amounts of fibrin, and the weight of fibrin producedby a given submaximal amount of thrombin is not affected by thetime during which the thrombin is allowed to act. I n any case,thrombin is not present as such in the blood, but in the form ofa precursor, prothrombin. For the conversion of prothrombininto thrombin there is needed yet another factor, thrombokinase orthromboplastin.This exerts its influence only in the presence ofcalcium salts. It is not present in the circulating blood, b u t issupplied a t the moment of shedding either from the leucocytes orfrom the extraneous tissues. I n the theory of Morawitz it is adynamic ‘‘ activator ” of prothrombin. On this view, the kinasepresumably acts as catalyst t o some chemical change. Its entryinto the plasma gives the initial impulse t o the changes whichfinally result in clot formation.The maintenance of fluidity in circulating blood is due, how-3 New York, William Wood and Company, 1915198 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.ever, not alone to the absence of kinase, but also to influenceswhich inhibit the action of thrombin.These inhibitory effectsare collectively attributed to “ antithrombin.” Howell treats of“ antithrombin” as a quite definite substance, and his theory ofcoagulation gives much prominence to it. According t o him, theconversion of prothrombin into thrombin calls for the action ofcalcium, and for that alone. This action does not occur i n circu-lating blood or in a stable plasma, because in them the pro-thrombin is combined with antithrombin. Throinbokinase(thromboplastin in Howell’s nomenclature) is not an “ activator ”in the previously accepted sense. It exerts its influence by com-bining with antithrombin, so liberating prothrombin from itscombination with t h a t substance and leaving it open to theinfluence of calcium.This view is apparently supported byHowell’s observation that if plasma is treated with acetone andthe resulting precipitate washed with ether, an aqueous solutionof the precipitate yields active thrombin on the addition of calciumalone. According to Howell,4 i t contains no kinase, because it willnot clot peptone plasma. It contains prothrombin which is opento the influence of lime salts, because it has been freed from anti-thrombin. H. H. Dale and G. s. Walpole,s however, in a paperwith which I am about to deal, state that they have hitherto failedto obtain on Howell’s lines a preparation with such properties,and they point out that, in any case, the proof of the absence ofkinase is unsatisfactory.These authors generally criticise the ex-periments upon which Howell bases his views, and supply whatseems to be very conclusive evidence to show t h a t there is no realantagonism between the actions of kinase and antithrombin, andtherefore t h a t Howell’s view as to the function of the former isunjustified.The paper of Dale and Walpole is one which marks real pro-gress in the subject. It describes improvements in technique whichyield a series of preparations such that each exhibits-uncompli-cated by any other action-the activities of some one of thefactors demanded by the theory of Morawitz. The mutualinfluence and relations of the various factors is thus made par-ticularly clear, and the views of Morawitz are confirmed.The interest’ of the paper is, however, by no means confined topoints of criticism or technique.It deals with the mode of originof thrombokinase, and in this connexion supplies facts of quiteunusual interest..I Amer. J . Physiol., 1014, 35, 474 ; A . , 191.5, i , 3.3.,i Bioclten?. .7., 1916, 10, 331 ; -4.. i , 859PHYSIOLOGICAL CHEMJ STRY. 199Recent work by others6 has shown that when serum is shakenwith chloroform it acquires remarkable toxic properties. Thisobservation, as a matter of fact, led to the present work of Daleand Walpole, whose concern was primarily with the study ofanaphylaxis. G. R. Minot7 has shown t h a t such treatment withchloroform destroys antithrombin. Dale and Walpole confirmthis, and find t h a t the destruction is both rapid and complete.When stable fowl’s plasma is shaken with twice its volume ofchloroform it clots, and the resulting serum is found to be richin free thrombin. Now, as the plasma loses its antithrombinunder this treatment, the result is, a t first sight, in favour ofHowell’s theory.The chloroform, in effect, seems to act as kinaseis, by him, assumed to act, liberating thrombin by the removal ofits inhibitor. It is found, however, that a minute proportion ofthe chloroform-treated plasma when mixed with fresh stable plasmawill cause this to clot with great rapidity, although it contains allits original supply of antithrombin. Moreover, chloroform doesnot yield free thrombin from such preparations of prothrombin as(on Howell’s view) would contain it in combination with anti-thrombin, although kinase activates these preparations a t once.Further experiments showed that, as a matter of fact, chloroforminitiates coagulation in a stable plasma, not because it destroysantithrombin, b u t because it induces the liberation of kinase,from which the plasma was previously free.This is certainly aninteresting result, and its explanation is still more interesting.Chloroform destroys not only the antithrombin, b u t also theantitryptic factor or factors in serum, as Jobling and hisco-workers 8 have shown ; trypsin is thus unmasked, and proteolyticactivity appears in the fluid. Now this fact suggests a possibleclue to the appearance of the thrombokinase in plasma. Dale andWalpole added trypsin under carefully controlled conditions t ostable bird’s plasma, and demonstrated completely t h a t as soon asi t is in excess of the antitrypsin, kinase is liberated. Chloroformthus causes plasma to coagulate, because by destroying antitrypsinit initiates proteolysis, which is followed by the liberation ofkinase. “It may be,” the authors write, “ t h a t some degree ofprotein cleavage is the necessary preliminary to the liberation ofkinase from any tissue: and that the ease with which it appearson cell injury and the relatively drastic treatment needed for itsproduction in a stable avian plasma represent a differing delicacyin the poise in the balance between proteolytic ferment and thoseJ .Exp. Med., 1914, 19, 482, with other papers Jobling and Peterson.in succeeding volumes.7 Amer.J . Physiol., 1915, 39, 131 ; A., i, 100. loc. cit200 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.antagonistic influences which, in the case of the plasma, we sum-marily describe as ‘ antitrypsin.’ ”Throinbokinase is certainly, from the biological point of view,a most interesting and important agent. Howell believes that itis identical with the phosphatide kephalin, and, under his direc-tion, J. McLean9 has recently sought for further evidence in sup-port of this view. H e prepared kephalin froin various sources,and, having purified his products as completely as possible, foundthat they all displayed the properties of kinase. This evidencegains in significance from the fact that other phosphatides werefound by McLean to be quite inactive.C. A. PekelharinglOhas also shown t h a t ordinary lecithins are without kinasic action.Recent work by P. A. Levene and C. J. Westll seems to showthat kephalin as obtained from quite diverse tissues is one andthe same substance, and, as its actual constitution is withinmeasurable distance of being accurately known, i t would seem thatin the identification of kinase with kephalin a real step forwardhas been made in the basal chemistry of blood coagulation. Whenwe remember, however, how stable may be the association betweencertain enzymes and lipoids, we may still feel t h a t a final proofof the identity will not come until kephalin has been synthesisedand the kinasic powers of the artificial product demonstrated.J.C. Wakelin Barrett12 has made a quantitative study of therelations which obtain between the time of coagulation of fibrinogenand the amount of thrombokinase concerned in the process. Heclaims t o have established the significant fact that a given quantityof kinase in the presence of calcium chloride and a sufficient amountof prothrombin will produce a definite quantity of thrombin whichis independent of the actual concentration of prothrombin. Hestates, moreover, although I do not fully appreciate the evidencefor the statement, that when a mixture of prothrombin, fibrinogen,and calcium salts coagulates under the influence of added kinase,the coagulation time observed is wholly occupied by the action ofthrombin on the fibrinogen.The time factor in the activation bykinase is negligible. These results suggest, perhaps, that theevents in activisation are physicochemical rather than chemical ;but they cannot a t present be held to support the view of Howell,as the possible influence of antithrombin factors was not consideredin Barrett’s work. His quantitative study of thrombin activitydoes not support Howell. He finds that when the concentration9 Amer. J . Physiol., 1916, 46, 250 ; A . , i, G93.l o Zeitsch. physiol. Chcm., 1914,89, 22 ; A . , 1914, i, 219.J . Biol. Chem., 1916,24, I11 ; A . , i, 298.Biochent. J . , 1916,Q. Fill ; A . , i , 229PHYSIOLOGICAL CHEMISTRY. 201of lime salts is properly adjusted, the coagulation process proceedsaccording t o the equationxy = nz,whenx =the time taken for complete coagulation,y =concentration of thrombin,5: =concentration of fibrinogen,71 = a constant.In a series of papers by E.Hekma13 we have the thesis sup-ported that the fibrin of blood clot is in the condition of areversible gel, which with proper treatment may be reconvertedinto a sol and again coagulated. Here again, apparently, is sup-port for the physicochemical conception of coagulation. Hekrna,l4and also Howell,l5 have studied fibrin formation with the ultra-microscope. They agree that it may first appear in the form ofultramicroscopic needles having the optical properties of crystals.No reference has yet been made in these Reports to the factthat adrenaline influences the coagulability of the blood.I f i tbe added to blood which subsequently traverses the liver andintestines, the organs are so stimulated as t o add something t othe blood which increases its coagulability. This property comesinto the picture of adrenal action as one specially directed t oprepare the animal f o r emergencies.16A naphyla xis.Between the agents which determine and control the events ofblood coagulation and those concerned in the phenomenon ofimmunity there are probably close relations. We cannot tell whichparticular line of study among all those which deal with thereaction of the animal to the intrusion of substances foreign t oits tissues is most likely t o lead to the illuminating generalisationswhich are being sought.It is justifiable to believe, however, thata proper understanding of the extraordinary facts of anaphylaxiswill make a real breach in our ignorance. Why does an in-finitesimal dose of a particular protein, when it escapes the pro-tective effects of digestion, so affect the tissues of an animal that,provided t,ime is given f o r certain events to mature, the receptionof a second dose becomes a fatal affair? What, in physicochemicall3 Biochem. Zeitsch., 1914,62, 161 ; 63, 204 ; 64, 86 ; A., 1914, i, 754, 875l4 Ibid., 1916,73, 370, 428 ; A . , i, 447, 448.l5 Amer. J . Physiol., 1914,25, 143 ; A., 1914, i, 1151.lG W. B. Cannon, W. L. Mendenhall, and H. Gray, ibid., 34,225 ; A . , 1914,1013.i , 765.H202 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.terms, does this “ sensitisation ” to particular proteins mean ?How comes it t h a t the reaction is a t times so rigidly specific?Something of fundamental importance t o biology resides in theanswers to these questions.The work recently described by H.H. Dale and P. Hartley17has shown t h a t when the three types of protein obtainable fromserum are carefully separated, each 0110 can act as a distinctanaphylactic antigen. It has shown, moreover, t h a t the sensitisa-tion may, in the case of successful preparations, be perfectlyspecific, so that euglobulin, for example, produces sensitiveness toeuglobulin and not to pseudoglobulin, whilst injection of albuminproduces a high degree of sensitiveness to albumin and none toglobulin.With pseudoglobulin preparations, such clear specificitywas never observed; but as this fraction is by far the most difficultto purify, there is little doubt t h a t when its injection sensitisesmore or less to the other fractions, it is merely because traces ofthese are still present. When it is remembered t h a t 1/20,000milligram of a particular protein may sensitise t o that protein,l*anaphylactic specificity may well be a very rigid test of the com-pleteness of a separation. These observations fully illustrate thefact, now becoming familiar, that specificity in an antigen is notdue to vague ‘ I biological’’ qualities, but to chemical or physico-chemical properties inherent in the antigen as a substance. Theblood of an individual yields two or more proteins which proveto be distinct antigens, whilst homologous proteins from quitedifferent species may prove indistinguishable.I n the paper underdiscussion it is shown, for example, t h a t the crystalline albuminsin ducks’ eggs a10 identical as antigens with those of the hen’segg. Such observations are not entirely new, but because of theexcellent technique which the authors employed they add greatlyto our confidence in the facts.We have nowgood evidence t h a t albumins are chemically distinct fromglobulins,’g but, so far as a balance in amino-acids is concerned,there would seem a t least t o be no difference between one class ofglobulin and another. It is, nevertheIess, difficult to believe thatantigenic distinctions could exist apart from the true chemicaldifferences, and, as Dale and Hartley remind us, there is abundantroom for such differences in the order of the linkings, even in thecase of two proteins with identical amino-acid content.20l7 Biochem.J., 1916, 10, 408 ; A., i, 859.l9 P. Hartley, Biochem. J., 1914,8, 541, A . , 1914, i, 1206 ; also C . Crowther2o H. W. Dudley and H. E. Woodman, ibid., 1015,9, 97 ; A . , 1915, i. 4G8.Upon what does antigenic specificity depend?H. G. Wells, J . Infect. Dis., 1908,5, 449.and H. Raistrick, ibid., 1916, 10, 434 ; A . , i, 864PHYSIOLOGICAL CHEMISTRY. 203The results of the research just discussed will be found to havepractical bearings in connexion with the concentration of antigensin the therapeutic sera.The hzmoglobins from the blood ofdifferent animals show marked specificity as anaphylacticantigens .21H. G. Wells 22 has shown that the so-called P-nucleoproteins,which are compounds of guanylic acid with protein, and can beextracted from organs with boiling water, possess definite antigenicproperties demonstrable by the anaphylaxis reaction. As thereare but few known proteins which retain their antigenic capacityafter boiling, the observation may give some indication as t o thenature of the protein moiety of these substances.Needless to say, many theories t o account for the facts ofanaphylaxis have already been advanced. I will refer very brieflyto a recent one which does not introduce any especially new con-ception, but will illustrate the trend of thought.It relates withanaphylactic phenomenon some of the newer facts which I havediscussed in connexion with blood coagulation.It was pointed out that treatment with chloroform not onlycauses plasma t o coagulate, but confers toxicity on serum. Thisaction is associated with a destruction of antitrypsin, and with aconsequent increase in the proteolytic power of the serum. It isconceivable, therefore, that the toxicity gained by the serum isdue to the formation of fission products from its proteins. Nowthe occurrence of an interaction between an antigen and its anti-substance affects the colloid equilibrium in circulating blood insuch a way as t o diminish its antitryptic properties. I n anaphy-laxis, the first, or sensitising, dose of a foreign protein produces ananti-substance, the excess of which continues t o circulate.Whenthe second dose of antigen is injected, the interaction with itsanti-substance occurs, and simultaneously the antitryptic powerof the blood is lowered. As a result, toxic fission products areproduced by the action of trypsin on the proteins of the blood,and these are responsible for the anaphylactic shock.This, stated very briefly, is the view put forward by J.Bronfenbrenner.23 The theory of this author has more originality,but is less easy t o follow, when he extends i t to bring the factsof passive immunity into relation with those of anaphylaxis. Imust not., however, give space to this extension here. There issome consonsus of opinion in favour of the view that i t is proteindisintegration products formed i n vivo which produce the remark-21 H.C. Bradley and W. D. Sansum, J . Biol. Ghem. 1914, 17, xsviii ; A . ,1914, i, 617. 22 Ibid., 1916,28, 11.23 Proc. SOC. expt. Biol. Med., New York, 1915,13, 19 ; A . , i, 181.H" 204 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.able symptoms of aiiaphylact.ic shock, although of their naturewe are as yet ignorant. On the other hand, it is generally heldthat only an intact protgiii can act as the antigen or sensitiser.The claim has recently been made24 that it is possible to renderrabbits anaphylactic by injecting so simple a substance as diglycyl-glycine. Several injections have to be made in order to sensitisethe animal.A final dose given after a longer interval than thosewhich intervened between the sensitising injections produced amoderate fall of blood pressure, with some acceleration of respira-tion and of intestinal movements. These symptoms are, it is true,part of the picture of anaphylactic shock. They are of very milddegree, however, and I think i t doubtful if the evidence given bythe experiments as a whole can be taken as showing that trueanaphylaxis can be produced by the tripeptide.The Hydrolysis of Proteins : Tissue Autolysis.Many details in the hydrolytic decomposition of proteins arestill obscure, and require to be cleared up for the sake of bothchemical and physiological knowledge. I n the case of acid hydro-lysis, for instance, destructive changes occur which are not yet fullyunderstood.A part of the nitrogen of the protein passes intothe so-called " humin " fraction, which consists of dark, amorphousproducts containing always an appreciable percentage of theoriginal protein nitrogen. These products call f o r study, becausethere has long been evidence that they are in some way related t onatural melanotic pigments, and also because their formationintroduces uncertainty into quantitative work on amino-acids asproduced by acid hydrolysis. It has been shown, for example,that any attempt to apply Van Slyke's amino-nitrogen determina-tion directly t o the analysis of feeding-stuffs breaks down, owingto the large and irregular yields of humins on hydrolysis.25The complete destruction of tryptophan which occurs whenprotein is boiled with acid turned attention to the possibility thatthe indole nucleus was concerned in humin formation.R. A.Gortner and M. J. Blish26 found that whilst zein yields very littlehumin when hydrolysed alone, a considerable amount is formedif tryptophan is added. The yield of humin from protein isalways increased when carbohydrates are present, and the authorsquoted shows that when tryptophan is mixed with carbohydratesand boiled with acid, nearly 90 per cent. of its nitrogen is24 E. Zunz and (Mlle.) Diakonoff, Biochem. J., 1916,10, 160 ; A , , i, 528.25 H. S. Grindley and M. E. Slnter, J. Amer. Chem. SOC., 1915,37, 2762 ; A . ,*' Ibid., 1630 ; A., 1915,i, 726.1915, ii, 598PHYSIOLOGlCAL CHEMISTRY.205recovered in “ humin.” They suggested that the reaction involvedin humin production was the condensation of an aldehyde withthe imino-group of the tryptophan nucleus, a type of reactionpreviously studied by Miss A. Homer.27 H. S. Grindley,28 how-ever, showed later that tryptophan-free zein increased its humin-nitrogen from 0-56 to 1-84 per cent. when hydrolysed in thepresence of dextrose, and suggested that other amino-acids besidestryptophan must share in nitrogenous humin formation. Gortner 2Qhas since admitted this. A recent paper by M. L. Roxas3O con-firms the fact, however, that tryptophan takes by far the largestshare in the process. When boiled with 20 per cent. hydrochloricacid plus sugar, the following amino-acids yielded humin, whilstthe proportion of their nitrogen concerned in the formation was :tryptophan, 71.0 ; tyrosine, 15 ; cystine, 3.1 ; arginine, 2-33 ; lysine,2.62; histidine, 1-84 per cent.I n the case of tryptophan, arginine,and histidine, the amino-group is concerned in the condensationwhich occurs, but with tyrosine and cystine another group seemsto be involved.Another point of practical interest t o protein studies is thequestion whether it is possible t o produce a complete liberationof all the amino-acids by the action of enzymes in uitro. Abder-halden has long claimed,31 and still claims,32 that by the successiveand prolonged action of pepsin, trypsin, and erepsin, protein iscompletely broken down. He has continued to use for feedingexperiments preparations so made, on the assumption that theyare free from polypeptides. This position has been challenged byHenriques33 and his co-workers.A. C. Andersen34 has recentlypublished papers bearing on the matter. According t o him, it isextremely difficult, if not impossible, to reach with the enzymes adegradation so complete that subsequent treatment with acid pro-duces no further changes in the products. He finds that aftervery prolonged zymolysis the end-products still yield, when boiledwith acids, a further quantity of free amino-groups and also a note-worthy proportion of extra free ammonia. The fact that thisammonia is liberated by acid, and not by enzymes, suggests that i tis derived from some special groupings in the protein molecule,and the author thinks that uramido-groups are its most probable27 Biochem. J ., 1913,7, 116 ; A . , 1913, ii, 451.z8 J . Amey. Ckem. SOC., 1916,37, 1778, 2762 ; z4., 1915, ii, 598; 1916, ii, 119.29 J . Biol. Chem., 1916, 26, 177 ; A . , i, 681.3 0 Ibid., 27, 71 ; A . , i, 797.31 See E. Abderlialderi and P. Rona, Zeitsch. p?iy/sioZ. C‘hem., 1907, 52, 508.33 Ihid., 1915,96 ; i, 147 ; A . , i, 580.33 Ibid., 1913,88, 361.34 Biochem. Zeitsch,., 1915, 70, 344 ; A., 1915, i. 1015206 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.Bource. E. M. Frankel35 has also published a recent elaboratestudy of the rate of decomposition of various proteins under theinfluence of the three digestive enzymes acting separately or con-secutively.He did not follow digestion to its final equilibrium,however, but he obtained by the consecutive action of all threeferments a liberation of as much as 85 to 90 per cent. of the amino-nitrogen in a period of seventeen to eighteen days. If, as Andersensuggests, there are really groups in the protein molecule which theknown enzymes of the digestive tract are unable to liberate, thepoint is one of interest. Otherwise the apparent differences in theexperience of different workers may be due merely to variationsin their technique.I n connexion with proteolysis by enzymes, reference must bemade to the remarkable statements of E. Herzfeld,36 who claimsto have shown t h a t the cleavage products of protein have them-selves proteolytic properties.According t o him, the activities ofpepsin preparations are largely due to proteoses and those oftrypsin to amino-acids, whilst even pure amino-acids have a pro-teolytic action.Now it has been shown by others that the typical curve of thevelocity of trypsin action differs from a logarithmic curve in t h a tit gradually falls away from such a curve owing to the retardinginfluence of the products of cleavage, especially amino-acids.37There is no evidence of autocatalysis such as Herzfeld’s claimswould call for. Indeed, the theoretical difficulties involved in hisstatements are such t h a t one is compelled t o look critically a t theexperimental evidence on which they are based. I n my opinionthis is not by any means convincing, and I am not surprised t ofind t h a t criticism is already beginning to make itself felt in the1iteratu1-e.~~ Some of the observations chronicled by Herzfeld call,nevertheless, for explanation, and his papers should receive atten-tion from other workers.It has recently been claimed by M Morse39 t h a t the degrada-tion of proteins in tissue autolysis is an autocatalytic phenomenon.A certain grade of acidity has long been known t o favour theprocess, and the author mentioned has found t h a t if the rate ofincrease of hydrion concentration which occurs during the auto-lysing of a tissue be plotted, it yields a curve of the logarithmictype, closely following the curve of proteolysis in the same tissue.35 J .Biol. Chem., 1916,26, 31 ; A . , i, 682.36 Biochem.Zeitsch., 1915, 68, 402, A., 1915, i , 468 ; ibid., 1915, 70, 262 ;37 W. M. Bayliss, Arch. Sci. Biol. Xt. Petersbzcrg, 1904, 11, Suppl. 281 ; A.,38 C. Funk, J . Biol. Chem. 1916,26, 121 ; L4., i, 767.39 J . Biol. Chem., 1916,24, 163: A . , i, 299.&4., 1915, i, 1019.1905. ii, 267PHYSIOLOGICAL CHEMISTRY. 207H e writes: “When we remember t h a t the development of acidityis the sine qua norn for autolysis, we have reason t o believe t h a t theprocess is an autocatalytic one, the developing acidity inducinggreater and greater acceleration in the digestion rate, and whenthe acid reaches a maximum in its production, digestion likewisereaches its maximum.” I n commenting on this paper, H. C.Bradley40 truly remarks that, since proteolysis is in question, theconclusion would really seem to affirm that the products of this-proteoses, amino-acids, etc.-accelerate the change, although,doubtless, this was not intended by Morse.It is clear, moreover,that none of the curves brought forward shows, as a matter offact, the slightest indication of acceleration in the rate of diges-tion. The term autocatalysis is therefore entirely misleading.True, the formation of acid within the tissue activates the auto-lytic enzymes, and in this sense the tissue, considered as a whole,initiates the degradation of its proteins; b u t when taking eventhis point of view, it has to be remembered that’ the acidity (dueto lactic acid) is not produceci in the process of proteolysis, buton quite other lines, so t h a t in no sense have we to deal withautocatalytic reactions.I n some experiments carried out by Bradley himself,41 and byBradley and Morse conjointly,42 the interesting fact was broughtto light that manganous chloride greatly accelerates liver auto-lysis.Moreover, whilst normally only 25 per cent. of the liverprotein is affected, in the presence of the salt 75 to 90 per cent.is digested. Of the total liver proteins, those from the connectivetissues are not attacked by the proteases present; hzmoglobin andalbumin are only attacked when accelerating agents are added,whilst globulins are readily and completely digested. Manganoussalts, it is suggested, so alter the resistant proteins as to renderthem vulnerable t o the ferments.Bradley 43 considers this viewto be confirmed when, in later work, he was able t o show t h a t anincrease in the concentration of substrate, brought about by theaddition of extraneous proteins, also produces acceleration. Thefact t h a t the liver proteases can digest foreign proteins was usedin an ingenious manner to throw light on the mechanism of auto-lysis. The digestion rate of the liver proteins themselves is closelyrelated to hydrion concentration, progressive increase beingobserved from PH=7.4 t o the optimum acidity (measured in adiaIysate) of PH=6.0. The rate of digestion of the foreign4 0 J . Biol. Chem., 1916,25, 201 ; A . , i, 582.41 Ibid., 1915, 20, xxix ; A . , 1916, i, 619.42 Ibid., 1915,21, 209 ; A . , 1915, i, 619.43 Ibid., 1915,22, 113 ; A., 1915, i, 1028208 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.proteins, caseinogen and peptone, for instance, is, however, notaffected by acid in a similar manner, and the effect of the acidcannot therefore be due t o an activation of enzymes.The true autolysis, involving the proteins of the organ itself,may also be almost inhibited by the addition of a salt, whichactually accelerates the degradation of added ~aseinogen.~~ Theaction of acids, including carbonic, is therefore on the substrate,not on the proteases.It is suggested that the liver cell itselfcontains two classes of proteins, namely: (1) available proteins orthose capable of being broken down by autolytic enzymes, and(2) resistant o r reserve proteins. The equilibrium between theseis dependent on the concentration of hydrogen ion in the cell a tany moment.A working hypothesis is based on these con-clusions and applied t o such occurrences in the body as involveautolysis and tissue destruction, or the reverse processes which occurin hypertrophy. Atrophy, involution, and necrosis are precededby a local asphyxia, which induces increase of acidity. Hyper-trophy is, as a rule, conditioned by an increased blood supply,which, by securing removal of carbon dioxide and the oxidation oforganic acids, as well as by constantly renewing the supply of thephosphate buffers from the blood, reduces autolysis to a minimum.The “available” proteins thus tend to be converted into thereserve type, and with increased synthesis of the former the cellproteins tend t o increase to a maximum. Simple hypertrophyresults.Such a contribution t o our conceptions of cell equilibriuinis, at any rate, worthy of attention.The Importance of Individual Amino-acids in Nutrition.This subject was dealt with in the lecture which I deliveredbefore the Chemical Society last May. When that lecture wasgiven, I was ignorant of the long paper Abderhalden publishedduring the year 1915. No abstract of this appeared in the Journalof the Chemical Society until August, 1916, and I had myselfhad no access to the journal in which it appeared.45 The paper,however, although comprising a discussion of considerable lengthon the special subject of this section, contributes but little to theactual facts.It contains further weighty evidence t o show thatthe animal is able to deal with the total amino-acids of completelypredigested protein, for i t describes the successful nutrition of adog f o r no fewer than 290 days on such material derived fromvarious sources. A second dog flourished on similar material44 H. C. Bradley and J. Taylor, J . Biol. Chem., 1916,25, 361. A . , i. 582.45 Zeitsch. physiol. Chem., 1915,96, 1 ; A . , i, 580PHYSIOLOGICAL CHEI\IISTRP. 309for 147 days. The experiments, on the other hand, in which thenutritive importance of individual amino-acids was tested were inmost cases inconclusive.That many workers should be independently engaged on astudy of this question is not surprising, for all must recognise itsinterest.Two methods of determining the nutritive importance of thisor that protein constituent have been used.The one consists inremoving individual amino-acids from the total products of hydro-lysis and feeding animals on the residuum. This was firstemployed, although with imperfect technique, by Henriques andWanson in 1905. Theother method has been employed in the admirable experiments ofOsborne and Mendel, by McCollum and Davis, and by some others.It consists in feeding with pure proteins in which, with some degreeof accuracy, the amino-acid balance is known. The effect onnutrition of deficiency or excess of a given constituent can be thusobserved. Each method has its advantages and its obviouslimitations.Another distinction in method must be mentioned.The nutri-tive effect may be observed in the gain o r loss of body-weight, orchanges in the nitrogen balance of the animal may be registered.The latter method has traditional support, and to some minds itrepresents the one indispensable test of nutritional success. Tothe study of some aspects of metabolism, information concerningnitrogenous equilibrium is, of course, essential, but in connexionwith experiments of the type now under discussion it is frequentlyunnecessary, and, as a test, i t may be, I think, even misleading.The objection to relying on body-weight changes is that they maynot concern the essential tissues, but represent merely gain or lossof fat, or even of water. No such criticism is jstifiable, however,if the experiment is reasonably prolonged.The fact that thenitrogen balance has been determined is sometimes made t o excusethe shortness of an experiment. I n such short studies, however,the determination may itself mislead. A negative balance inducedby the omission of a constituent from the diet after a controlperiod showing equilibrium may, if of a comparatively brief dura-tion, represent no more than a period of adjustment. The nitrogenbalance is a safe indication of quantitative deficiencies in the totalenergy supply, but unless it be follotved for considerable periodsit is a less safe guide as to the real and permanent importance ofsome qualitative change in the diet. I think that this objectionapplies to certain of Abderhalden’s experiments.It should bepointed out, however, that in the paper under notice he himselfIt has been largely used by Abderhalden210 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.repudiates a blind trust in the factor of nitrogenous equilibrium.46Abderhalden contributes further evidence to show that tryptophanis indispensable, a fact on which all observers are agreed. Hefinds that the (not wholly complete) removal of tyrosine producesa t once a negative nitrogen balance in the animal, but that phenyl-alanine can, at least partly, replace the tyrosine. I n this con-nexion, it is to be noted that G. Totani47 obtained prolongedmaintenance of his animals, and even some growth, after removingtyrosine as completely as possible.Even when hydrolysed gelatinwith added tryptophan was given, maintenance was satisfactorilyaccomplished. Totani concluded that the functions of tyrosine werebeing subserved by phenylalanine, and believed, further, that theanimal has some power of synthesising orthocyclic amino-acids.H. H. Mitchell 48 has also published experiments suggesting thatthe orthocyclic acids are not altogether indispensable. Withregard t o other amino-acids, Abderhalden’s experiments-at leastup to the date of the 1915 paper now under notice-supply nosatisfactory information. He found, for instance, that the removalof lysine, of arginine, o r of histidine left the amino-acid mixturewithout any power of maintenance, but unfortunately ir? none ofthese cases was that power restored when the missing amino-acidwas replaced. From this circumstance, it was concluded that theprocess of removal had in some way altered the nature of one orother of the remaining constituents. This was not a t all theexperience of H.Ackroyd and myself.49 By a careful applica-tion of Kossel and Kutscher’s method to the hydrolysis mixture,we removed arginine and histidine together, and afterwardsreplaced one or both. It was easy to show that in the absenceof both there was nutritional failure, and restoration when bothwere replaced. I n our experiments, however, the fact came t olight that either one of these two protein constituents can to alarge degree serve vicariously for the other.Abderhalden draws the interesting conclusion that proline is notindispensable.I n the case of cystine and glutamic acid, theonly other units investigated, his results were inconclusive. Withregard to the former of these two, there can be little doubt, how-ever, that it is essential, not only because it carries the sulphur ofprotein, but from general evidence already available. F. W. Fore-man and I s o have obtained results which at least suggest thatglutamic and aspartic acids can be dispensed with.48 loc. cit., p. 51.47 Biochem. J . , 1!316,10, 382 ; A . , i , 860.48 J . Biol. Chem., 1916,26, 231 ; A . , i, 690.4 9 Paper in the press ; Biochem. J., 10,5 0 Lecture ; T., 1916, 109, 639PHYSIOLOGICAL CHEMISTRY. 211Studies based on the removal of individual amino-acids fromthe food are only now beginning; I am sure they will ultimatelyyield very valuable information.I n observations described in the paper already referred to,H.H. Mitchell adopted the plan of feeding with mixtures of pureamino-acids. Such diets are, of course, very costly, and a t thepresent time impossible t o prepare in considerable quantities. Theexperiments were theref ore confined to mice. The chief conclusionindicated, apart from that concerning the relatively small import-ance of tyrosine, is that, except in the case of preparations freefrom tryptophan, there is little difference-as judged by the rateof decline in body-weight and the survival period-in the nutri-tive effect of mixtures of very varying composition. The mixturessupplied were, of course, never complete, and their amount wasusually quantitatively insufficient.I n a general discussion, thisauthor refers to the “suggestive theory of Osborne and Mendelto the effect t h a t the only reason for the destruction of anyprotein a t all in maintenance might, be to liberate a small amountof one or two, o r a t most a relatively small number, of amino-acidsto engage in some ‘ hormone-like physiological duty upon whichproper metabolism might depend,’ ” whilst the view t h a t “ a t leastsome of the amino-acids have specific functions in metabolism, asidefrom that of serving simply as material for the synthesis of bodyprotein,” is referred to as ‘(recently and tentatively p u t forwardby Osborne and Mendel.” I may be permitted to point out thatthis theory, and no less the view, were explicitly put forward morethan ten years ag0.51Osborne and Mendel j2 have brought forward further evidenceto show t h a t the nutritive value of lactalbumin is superior to t h a tof caseinogen, a circumstance of practical importance, dependingdoubtless on the balance of amino-acids characteristic of theseproteins.When the protein supply was cut down to a minimum, it wasfound t h a t to produce the same gain in body-weight “50 per cent.more casein than lactalbumh was required, and of edestin nearly90 per cent.more.” I n other experiments it was found t h a t withapproximately equal intakes of total food there was an equal gainof weight when lactalbumin formed 8, casein 12, and edestin 15per cent.of the dietary. The addition of cystine, equivalent t o3 per cent. of the casein employed, made t h a t protein much moreefficient. The same authors have also published further experi-ments on the effects of adding tryptophan and lysine to zein. The51 F. G. Hopkins, Scieizce Progress, New Series, Vol. I, No. 1, July, 1906 ;Miss E. G. Willcock and I?. G. Hopkins, J . Physiol., 1906,35, 88.52 J . Biol. Chem., 1916,26, 1 ; A . , i, 690.Edestin is inferior t o either of the milk proteins212 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRYanimal is maintained when the former is added alone, althoughnot with the latter alone. When both are added there is growth.The quantities of these two amino-acids can be made the limitingfactors which determine nutritive equilibrium and the possibilityof growth.They afford an illustration of the “law of theminimum ” as applied to the food supply of an animal.That lysine is necessary for growth is shown in the experimentsof others.53 It seems to me entirely wrong, however, to speak ofthis diamino-acid as though it were a special growth factor or as“stimulating” growth. It is, like tryptophan and some otherprotein units, material which the animal cannot synthesis0 foritself, and, being a tissue constituent, it must therefore be presentin the food.The Residual Nitrogen of t h e Blood.I. Bang has published a long series of papers on the residual,or non-protein, nitrogen in the circulating blood of man and ofvarious animals. Some of the results given are not altogether inaccordance with those obtained by others, and some are puzzling,but the work of so experienced an experimentalist on so importanta subject calls for attention.The methods used were the author’s own.I am unable t oappraise them, as they have been described only in a monographpublished a t Wiesbaden during the war. The urea nitrogenand the amino-acid nitrogen were separately determined. Theformer varies between 6 and 20 milligrams per C.C. of blood, witha mean of 12 milligrams; the latter between 3 and 22 milligrams,also with a mean of 12 milligrams.54 These figures apply to humanblood and t o that of the ox, sheep, pig, and horse, all of whichagree closely. Those for the rabbit are higher. I n starvation,55the urea unexpectedly rises, but this effect is largely due t odeficiency of water, and perhaps to some renal factor.Starvatior,has no effect on the amino-acid nitrogen, a circumstance which,although a t first surprising, is in accordance with the observationsof others. It indicates that there is a regulatory mechanism intissue autolysis which maintains a definite equilibrium of circu-Iating amino-acids. More remarkable and less in accordance withthe statements of others is Bang’s claim that a large meal ofprotein does not increase the amino-acid nitrogen, although it pro-53 G. D. Buckner, E.H. Nollau,and J. I T . Kastle, -4mer. J . Physiol., 1915, 39,162 ; -4., i, 102.5 4 Biochem. Zeitsch., 1915, 72, 104 ; A . . i. 178.8 5 Ihid., 119 ; A., i, 178PHYSIOLOGICAL CHEMISTRY. 213duces a brief increase in the urea.When free amino-acids aregiven’s6 an increase is observed, especially in fasting animals, aboutan hour after the administration. So far, the observations arereconcilable with current views concerning the digestion andabsorption of proteins. Work was quoted in last year’s Report toshow how rapid are the processes of intermediate metabolism, andi t may well be, in special circumstances a t any rate, that with thecomparatively slow absorption which goes on during normal diges-tion the amino-acids are dealt with rapidly enough to prevent anyappreciable rise in their concentration in the blood. This neednot remain true with the more rapid absorption of free amino-acids when directly administered.Much more puzzling, however, are results which seem to showthat a diminution of amino-acids in the blood, and even theirdisappearance therefrom, may occur after a large protein meal.There seems to be no explanation for this, and if Bang’s observa-tions are confirmed, we have an obscure phenomenon before US.I find certain other results57 reported by him equally difficult tounderstand. As a consequence of giving glycine by the mouth t orabbits, an increase of amino-acid nitrogen in the blood wasobserved which, within certain limits of administration, wasroughly proportional to the amount of glycine ingested. Equi-valent quantities of leucine, however, produced no such increase.Its administration was followed, on the other hand, by a rapidrise in the urea of the urine.Leucine is more rapidlydeaminised than glycine. These observations led t o experimentsin which a mixture of the above amino-acids, and also gelatin andcertain proteins, were given. From the results as a whole, theconclusion is drawn that those proteins which on hydrolysis yieldlittle or no glycine give rise on their ingestion to no increase inthe amino-acids of the blood, whereas those yielding relativelylarge amounts cause a corresponding increase in the circulatoryamino-acid. It is not difficult to understand that the acids ofhigher molecular weight may be more readily deaminised in theliver than is glycine. It is most difficult, however, to believe that,under normal conditions of absorption and metabolism, glycinealone among amino-acids should pass the liver unchanged.It ismore easy t o believe that glycine-which is but a small constituentof any tissue proteins save those of the stable connective tissues-is less rapidly dealt with by the tissues as a whole, and thereforeremains longer in the blood.That the tissues offer differential treatment to the various56 Biochem. Zeitsch., 1915, 72, 129 ; A , , i, 178.I b i d . , 1916,74, 278 ; A., i, 579214 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.amino-acids is suggested by some experiments carried out byU. Lombroso.58 I n these, amino-acids, dissolved either in bloodor in Ringer’s solution, were perfused through isolated survivingorgans. It was found, in agreement with Bang’s results, t h a tglycine was but little affected during passage through the liver.It was broken down in muscle, however.Alanine, on the otherhand, is apparently less readily dealt with in the muscle than incertain other organs. The experinients of this author furtherindicate t h a t an important difference exists between blood andsalt solutions as perfusion media. When Ringer’s solution wasused, the disappearance of amino-acids during perfusion wasassociated with evidence of deamination and oxidation. Whenblood was employed there was, in the opinion of the author, moreevidence for the occurrence of synthesis.The s!ig,o1, of t h e Blood.Some three years ago I . Bang published an account of h kresearches on blood sugar in the form of a separate monograph.59Among matters of interest brought forward in the book wasevidence t o show that the liver itself deals with a smaller propor-tion of the sugar ingested a t any time than we have been apt tosuppose.Estimations in the blood showed that during the absorp-tion of a carbohydrate meal a large amount of sugar escapes theliver. It would seem that we must credit the tissues as a wholewith considerable power of storing carbohydrate in one form oranother. G. Graham60 has supported such a view with the resultsof experiments carried out on himself. H e makes the interest-ing suggestion that when the body is fatigued there is a lessenedability to store carbohydrate, so that under conditions of fatiguethe rise of blood sugar which results from a given ingestion isconsiderably above the normal.U. Lombroso61 has investigated the fate of blood sugar underexperimental conditions which are more or less new.Some of hisresults call for critical attention. H e maintains t h a t an explana-tion for the contradictory results which have been reported con-cerning the fate of sugar in she& blood is to be found in thecircumstance that a synthetic conciensation of the dextrose in theblood, as revealed by the experiments of Levene and Meyer, occurs5 8 Atti R. Accad. Lincei, 1915, [v], 24, ii, 401 ; A., i. 101 : compare ihid.,5 9 “ Der Blutzucker,” WiesFaden, 1913.6 0 J . Physiol., 1916,50, 285 ; A., i, 613.61 Atti R. Accad. Lincei, 1916, [v], 25, i , 736 ; A . , i, 612 ; ibid., 802 ; /I.,24.i, 1253; A ., 1915, j , 917.i , 686 ; i h i d . , ii, 41 ; A . , i, 686 ; ibid., 83 ; A . , i, 770PHYSIOLOGICAL C HEM1 STRY. 21 5a t the outset, but if the blood is allowed to remain longer, this isfollowed by a degradation of the complex, and subsequent destruc-tion of the liberated sugar. The synthetic function and the glyco-lytic function are independent variables. If blood with addedsugar is circulated through the pancreas or intestine, there is anincrease in its glycolytic power, with little or no change in itspower to condense the sugar. If, on the other hand, fresh bloodpoor in sugar is first circuiated through the same organs, andsugar is subsequently added to it, a marked increase in the syntheticpower is observed, the glycolytic power being scarcely affected.The blood of a depancreatised dog has little glycolytic capacity,but its power to condense sugar remains almost normal.@ If it iscirculated through the intestine of the animal from which it wastaken, it gains greatly in the glycolytic factor.Clearly the moststriking of this author’s claims are those relating to the influenceof the intestine. The gut yields a glycolytic agent, or, in othercircumstances, a condensing zgent, to the blood. It yields themeven more readily than the pancreas. Some of these results seemto me quite inexplicable a t present, b u t that’ the intestinal epi-thelium might have to be classed among the tissues specially con-cerned in carbohydrate equilibrium has always been probable.R ennl Perm en bilit y.The present availability of methods which permit of the estima-tion of blood constituents in very small quantities of material hasencouraged activity in various fields.Many, for instance, havebeen engaged on investigations concerning the permeability ofthe kidney. Physiologists want to know how far the excretion ofthis or that urinary constituent is dependent on its concentra-tion in the blood, and whether the rate of excretion is controlledby differences of concentration in the blood and urine respectively.Does the normal kidney display a constant permeability ” whensuch factors are constant? I f so, and if a measure of normalpermeability can be established, pathologists require to know howthis permeability varies in disease. This is, of course, very farfrom being a new region for research, but work on it has gainedimpetus now t h a t accurate estimations can be made in the circu-lating blood without inconvenience to the subject of experiment.A few years ago Ambard and Wei163 asserted t h a t in the caseof urea such relations hold t h a t with a given concentration in theurine the rate of its secretion varies as the square of its concen-62 Compare H.McLcan. J . PhysioZ., 1916, 50. 16s ; A . , i, 61 3.G3 J . Physiol. et Path. gek., 1913. 14, 753216 ANNUAL REPOEL’I’S ON TEIE PROGRESS O F CI-IEMIS‘I’HY.tration in the blood, whilst with a given concentration in the bloodthe rate varies as the square root of the urinary concentration.These relations, as obtained by experiment, they combined in aformula, which is mostly used in the form:Concentration in the blood =constant,dRate of excretionThey found this t o hold in a great number of normal people undervery varying conditions.Any considerable departure from it wastaken as an indication of renal abnormality. These claims havebeen supported by 0thers.6~During the present year, however, T. Addis and C. K. Wata-nabe65 have criticised Ambard and Weil’s law of excretion. Itis common to find in physiology that, once a “law” is promul-gated, its limitations begin t o come t o light. The authors justquoted find that factors other than the concentration of urea areimportant in determining the rate of its excretion by the normalkidney.They found, it is true, in thirty-nine experiments doneupon young healthy adults that when the conditions are the samethe organ displays a high degree of constancy of function.66 Largedoses (24 and 40 grams respectively in different experiments) ofurea were given by the mouth, and the rate of its excretionfollowed in successive periods after the administration. This ratewas remarkably constant both when different individuals were com-pared and also in the same individual on different occasions. Theexcretion rate, however, was in each case lower in the earliesthours of the experiment, when the concentration in the blood wasdirectly shown to be a t its highest, than in the immediately succeed-ing periods with less in the blood.The kidney is not passive.The successive administration of identical doses of urea proved,from the identity of the excretion rate following each dose,that “kidney fatigue ” could not be induced under the conditionsof these experiments. This leads me to refer once more to one ofI. Bang’s67 papers. This author, finding that after givingurea to rabbits there was a somewhat prolonged display ofexcess in the blood, came t o the conclusion that the kidney wasfatigued by the extra calls made on it. He gave very large64 Compare F. C. McLean and L. Selling, J . BioZ. Chem., 1914, 19, 31 ; A . ,1914, i, 1183. Also F. C. McLean, Amer. J . Physiol., 1915, 36, 357 ; A . , 1915,i, 186.65 J . Biol. Chem., 1916,24, 203 ; A., i, 352.66 Ibid., 27, 249 ; A., i, 864.G7 Biochem.Zeitsch., 1915,72, 139 ; A . , 1915, i, 179PHYSIOLOGICAL CHEMISTRY. 217doses, however, apparently about 5 grams per kilo., and for thisreason his results are less important in their bearing on thevery interesting question of physiological renal fatigue. His doseswere large enough, indeed, t o prove very poisonous t o the animals.H e ascribes the effects to ammonia poisoning, having found t h a tammonia is largely increased in the blood as the result of givingurea. This observation is interesting in its suggestion thatthe reaction by which urea normally is formed fromammonium salts is reversible. It is not easy, however, to reconcileBang's findings with certain other experimental facts, f o r B. C. P.Jansen 68 has recently confirmed the earlier results of Wakemanand DakinGQ in showing that perfusing the liver with blood con-taining ten times the normal concentration of urea leads to noammonia formation. The normal reaction by which urea isformed from ammonium salts seems, therefore, to be irreversibleunder these conditions. I n parenthesis, I may point out t.hatalthough creatinine is formed from the creatine of muscle, W. C.Rose and F. W. Dimmitt7O have shown that a large increase of theformer in the blood leads to no increase of the latter in the muscles.This reaction again seems to be irreversible. No one, I think, hasyet observed the reversal of either of these reactions under thecontrol of suitable enzymes. It would be most interesting t oknow in every such case how far the apparent irreversibility isinherent in the reaction and how far the properties of the livingcell intrude in the matter.Having regard to the results just quoted, it is difficult to under-stand how and where the large ammonia formation from circu-lating urea could have occurred in Bang's experiments.Returning to the subject with which this section is more par-ticularly concerned, I may direct attention to the interestingobservations of V. C. Myers, M. S. Fine, and W. G. Lough,71which show t h a t whilst in the early stages of clinical nephritisuric acid is the excretive which chiefly accumulates in the blood,a t later periods, when the permeability of the kidney has furtherdecreased, urea accumulates and then creatinine. When the diseasenears the fatal stage, the amount of the last-mentioned excretivefound in the blood may be twenty times the normal. Severenephritis may markedly reduce the permeability of the kidney todextrose and upset any normal relation between the sugar in theArch. Neerland, 1915 [iii B], 2, 594 ; A., i, 299.G 9 J . Biol. Chem., 1911, 9, 327 ; A . , 1911, ii, 629.' O J . Biol. Chem., 1916,26, 345 ; A . , i, 774.' 1 Proc. Xoc. expt. Riol. Med., New York, 1915, 13, 5 ; .1., i. 193218 ANNUAL REPORTS ON THE PROGRESS OF CHEMlSTRY.blood and urine. I n diabetes, for instance, when the disease iscomplicated by nephritis, a given concentration of sugar in theurine may correspond with a much higher concentration in theblood than i t would indicate in a case free from such renal changes.The renal permeability for sugar may vary considerably, however,without the existence of actual nephritis, as many papers dealingwith the raising or lowering of the “leak-point ” in diabetes haveshown .zF. GOWLAND HOPKINS.72 See G. Graham, Proc. p h p i o ? . Soc., 1916, xlvi. ; L4., i, 1915, i, 1036

ISSN:0365-6217    

DOI:10.1039/AR9161300195

出版商:RSC    

年代:1916

数据来源: RSC

摘要:

AGRICULTURAL CHEMISTRY AND VEGETABLEPHYSIOLOGY.ALTHOUGH there has been a decided falling off in the rate ofpublication of some of the Continental journals, the number ofpapers published in 1916 does not show any very considerablereduction, owing, to a great extent, to increased activity in theUnited States.The obituary list includes the well-known name of E. W. Hilgard,who for more than forty years held the positions of Professor ofAgriculture and Director of the Experiment Station of California.Among new publications may be mentioned a monthly journal,Soil Science, edited by J. G. Lipman ; " Organic AgriculturalChemistry," by 5. C. Chamberlain ; " Beitrage zur Kenntniss derErnahrung der Zuckerrube," by J. Stoklasa and A. Matousek; and'' Die Kalirohsalze, ihre Gewinnung und Verarbeitung," byW.Michels and C. Przibylla.The 3 tniosyhere.The Committee for the Investigation of Atmospheric Pollutionhave issued their first report,l for the year ending March, 1915,which contains the results of monthly analyses of rain-water madea t twelve places for a whole year, and a t a number of other placesfor shorter periods. The positions selected were, naturally, mainlyin the largest towns. Malvern and Exeter were, however, included,and a t these places it was found that the nitrogen in the form ofammonia amounted to only 1.9 kilograms per hectare per annum,the lowest yearly result hitherto recorded in England. A tMalvern, the sulphates are unexpectedly high, the yearly amountper hectare being 50.5 kilograms, as compared with 19.5 kilogramsa t Rothamsted.The analyses are very complete, everything ofimportance being included, with the exception of nitrates, theexclusion of which, no doubt unavoidable, is to be regretted, asthe figures would have been a usefnl check on the ammonia results.Lancet, Feb. 26, 1916, 190, Suppl. ; A . , i, 592."1'220 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.I n connexion with the results found a t Sheffield, it has beenpointed out 2 that the contamination of the air is a t its worst oncalm, foggy days when no rain is falling, and that on such occasionsthe method employed fails t o give any record. That, however,will depend on the length of the periods without rain and on otheratmospheric conditions.If the fog is blown away before the raincomes, the impurities which it contains will, of course, be lost; thefog may, however, be followed by showers, in which case a fairlygood indication of the extent of the contamination will be obtained.I n rainy weather, the greater part of the smoke (much of whichis rapidly distributed over the surrounding country in fineweather) will be brought down by the rain as soon as it leavesthe chimneys, so that, with the rainfalls of the various districtsconcerned, it is perhaps more likely t h a t the rain-water resultsindicate too much contamination rather than too little.Analyses of the air itself would add considerably to the valueof the rain-water results, even if only three or four constituentswere estimated.In California,3 an investigation of the atmosphere, in whichnearly five thousand estimations of sulphur dioxide were made a ttwo stations in a smelter zone, showed t h a t the air may, underexceptional conditions, contain as much as 7.1 per million, and t h a teven in San Francisco the amount’ of sulphur dioxide may reachnearly 2 per million.The average amount, calculated from allthe observations, was, however, only 0.22 per million, whilst theaverage, under conditions favourable for high contamination, was3.6 per million.Analyses of rain-water have also been made a t Mount Vernon,I o w ~ , ~ a t St. Croix,5 and a t Montevideo.6Soils.A new kind of clay, having unusual properties, has been dis-covered in some unreclaimed land near the city of Mexico.’ Afew grams of the soil, which looks like an ordinary dark grey clay,placed in a 50 C.C.cylinder with some water, swelled very rapidly,and was found the next day completely t o fill the cylinder. Ananalysis of the clay showed that it contains very little aluminiumW. P. Wynne, “Rep. to the Health Committee of the Sheffield CityCouncil,” 1914-15.J. A. Holmes, E. C . Franklin, and R. A. Gould, U.S. Bur. Mines, Bzrll.R. Artis, Chem. News, 1916, 113, 3 ; A . , i, 304.Schroeder, Exper. Stat. Record, 1916, 34, 16.98, 1916.ii L. Smith, Rep. Agric. Exper. Stat. 8t. Croix, Copeiihagcn, 1915.’i E. W. Hilgard, Proc. Nut. Acad. Sci., 1916, 2, 8AGRICULTURAL CHEMISTRY BND VEGETABLE PHTSZOLOGY. 221in relation to the silica, and that the predominating base is mag-nesium.The soil, which was originally supposed to be afflictedwith " alkali," contains comparatively small amounts of sodiumsulphate and carbonate; it contains a good deal of calciumcarbonate, and cannot therefore be reclaimed by liming, so thatacid treatment would seem to be the only possible effective agent.It is suggested that the material may be of practical use inclosing the crevices in oil wells which are liable to flows of water.Experiments on the flocculation of soil colloidal solutions showedthat the effect of the same electrolyte varied considerably accord-ing to the composition of the soils, solutions from clay soils floccu-lating most readily, then loam soils, and lastly peat.* Quanti-tative experiments, in which the different soil solutions werebrought to the same concentration (100 C.C.= 0.027 gram), showedthat lead salts acted better than ferric sulphate and aluminiumpotassium sulphate, and that the minimum amount of electrolyterequired was nearly the same with the best coagulants, lead nitrateand acetate and mercurous nitrate, whilst with other salts theamounts were very variable. Organic matter has a considerableeffect on the stability of the colloid solution, five times as muchNI5-nitric acid and seven times as much saturated calcium hydr.oxide solution being required in the case of peat as with clay,whilst with a mixture of clay and peat there is a regular rise inthe amount of electrolyte required as the proportion of peat toclay is increased.Colloid solutions when diluted with waterrequire less electrolyte for flocculation; the decrease is not, how-ever, in proportion t o the decrease in concentration.Recent results in connexion with the question of the effect ofclimate on soils showed that Kansas and Maryland soils, whentransferred to California for five years, became more deeply redin colour, and that Californian and Kansas soils were bleached t olight grey or yellowish-grey in Maryland.9 Samples of any onesoil from the three stations were so changed as to become unrecog-nisable, and seemed to represent three very distinct types. As arule, it was found that the numbers of bacteria, ammonification,nitrification, and fixation of nitrogen increased when arid soilswere placed under humid conditions, and vice versa.The decom-position of cellulose, on the other hand, was always more rapidunder arid than under humid conditions. Changes in compositionwere also observed; it was found that under arid conditions thereis a tendency for iron to increase and for aluminium t o decrease,whilst the change is reversed when Californian soils are transferredM. I. Wolkoff, Soil S c i . , 1916, I, 585 ; A., i, 784.C. B. Lipman and D. D. Waynick, ihid., 6 222 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.to Kansas. Even without being moved, a soil may undergo coil-siderable change in five years; in a Maryland soil the amount ofmagnesia was reduced from 2.3 t o 0.21 per cent.The increased loss of carbon under arid conditions would partlyaccount for the decreased activity of the nitrifying and nitrogen-fixing organisms, and the former would be injuriously affected byunabsorbed ammonia and by organic matter dissolved by ammonia.The fixation of phosphoric acid, dissolved from soils by diluteacids, which sometimes takes place when the extracts are left incontact with the soils, has been found t o vary in extent accordingto the acid employed.10 This might be explained on the assump-tion that the acid slowly decomposes aluminium or iron com-pounds, producing soluble salts, which react with the dissolvedphosphoric acid; and the fact that with citric acid as the solventemployed the precipitation of phosphoric acid is less than withmineral acids would seem to support this view.It is, however,shown that not only phosphoric acid is taken up, but that oxalicacid is also withdrawn from its solution when left for a short timein contact with a soil in the presence of nitric acid. With 25 gramsof a subsoil, 1 litre of N/20-nitric acid, and 0.094 gram of oxalicacid, about 82 per cent. of the oxalic acid was rendered insoluble,whilst with ten times as much oxalic acid about 25 per cent. waswithdrawn from the solution.11 Both oxalic and citric acids aremore readily absorbed than phosphoric acid, whilst sulphuric acidis less absorbed, and hydrochloric and nitric acids not at all, or, atany rate, not appreciably. The withdrawal of acids from theirsolutions is attributed t o adsorption by soil colloids, and is shownto follow the usual law.The order of the extent of adsorptionof the different acids by soils is shown to agree with results obtainedby Skraup with filter paper. The variations in the amounts ofphosphoric acid taken up in presence of the different acids seemsto depend on the relative adsorption of the acid solvent, so thatwith citric acid, which is readily adsorbed, the withdrawal of phos-phoric acid is less than with nitric and hydrochloric acids, whichdo not seem t o be adsorbed a t all.Until more is known as t o the behaviour of plants towardsadsorbed substances, i t cannot definitely be said whether the avail-ability of phosphoric acid in soils is more correctly indicated byresults obtained after a short period of extraction or after moreprolonged extraction-presumably the former, as phosphoric acidliberated by root action would probably not remain long enoughin the soil to be adsorbed to any extent.This would correspondlo E. J. Russell and J. A. Yrescott, J . Ayric. Sci., 101G,8, 65.l1 J. A. Prescott, ibid., 111AGRICULTURAL CHEMISTRY AND VEGETABLE PHYSIOLOGY. 223with a diffusion method, which was found to give almost identicalresults with three different acidsSome experiments in which a sandy loam was treated with asolution of potassium chloride (approximately N/500), and thenleached with water, showed t h a t the process of adsorption may bereversed, the water being more rapidly adsorbed than the dissolvedsalt, so that the concentration of the percolates was increased ascompared with the original solution.121nvestigat.ions on the solutions present in soils have usuallybeen made with drainage or with soil extracts, which involve con-siderable dilution of the water originally present. A method hasnow been employed in which the soil water is separated from thesoil by a pressure of 300 kilograms or more per sq.cm. I n thisway, sufficient amounts of solution can be obtained from soils con-taining considerable amounts of clay or organic matter.l3Analyses of some expressed solutions showed that the amount ofcalcium varies a good deal in the surface soil, but remains constantin the subsoil for a considerable part of the year. The amountsof potassium were always found to be lower in unmanured than inmanured soils, and the movements of potassium were found t o beanalogous to those of calcium.The concentration of the soil solu-tion, to a depth of 50 cm., is diminished by heavy rainfalls andincreased by prolonged evaporation. Whilst the variations in thecourse of a year are considerable, the curves for potassium andcalcium are remarkably similar, and seem rather t o p u t out ofcourt the accepted ideas with regard to the absorption of potassium.Other experiments with expressed soil solutions have also givenindications t h a t the method is likely t o be of value.14 It has alsobeen found that, in physical analyses of soils, the influence ofmanures and of season are clearly indicated when, instead of dis-tilled water, a soil solution is employed.For this purpose, anartificial solution has t o be used, owing to the large amountsrequired.Analyses of drainage waters from lysimeter tanks, extendingover five years,15 have showed that applications of lime t o the soilfailed t o liberate potassium, and crop results indicated t h a tassimilation of potassium was not increased by lime. The amountof magnesium in the drainage was, however, increased ; applica-tions of potassium sulphate increased the calcium and magnesium,but not the potassium.l2 A. G. McCall, F. M. Hildebrandt, and E. S. Johnston, J . Physical ChPnz.l3 E. Ramann, S. Marz, and H. Bnuer, Internat. Mitt. Bodenk., 1918, 6 , 1 .l4 J. P. van Zyl, J . Landw., 1916, 64, 201.l5 T. L. Lyon and J. A. Bizzell, J.Amer. Soc. Agron., 1916,8, 81.1916, 20, 51 ; -4., i, 304As regards sulphur, it was found t h a t the unlimed soil, whichhad received some farmyard manure, lost 49 kilos. of sulphur perhectare, and that the limed soil lost a greater amount. The soilswhich received potassium sulphate every year lost from one-halfto one-third of the sulphur in the drainage.So,il A c idi t y .An investigation of some unsatisfactory soils near Jena showedt h a t they were more or less acid, owing, probably, to the actionof a superficial layer of raw humus in former times.16 Some ofthe soils which are still covered with humus are more stronglyacid, and these react with solutions of neutral salts. As a rule,the power of reacting with potassium chloride belongs chiefly tothe subsoils and to the uncultivated portions, so t h a t cultivationseems to remove the kind of acidity which, according to Daikuhara,depends on replacement of iron and aluminium by stronger basesand the hydrolysis of the resulting iron and aluminium salts.Most of the cultivated, acid surface soils reacted only with sodiumand calcium acetate.I n order to throw further light on the process by which acidityis produced, other soils from pine forests were investigated.Thesoils were covered with layers of raw humus from 1 to 3 cm. thick;all of them reacted with potassium chloride, and when boiled withnormal solutions for an hour the filtrates showed strong aciditywhen titrated with sodium hydroxide, and gave white precipitatesconsisting mainly of aluminium hydroxide.The acidity, whichdecreased with the depth a t which the samples were taken, dis-appeared a t a depth of 1.5 metres.As regards the action of humus in producing soluble aluminiumand iron salts in soils, it is shown that the true acidity of humusextracts is approximately the same as t h a t of acetic acid solutionsof the same strength, and that mineral soils when treated withextracts of humus render the soils capable of reacting withpotassium chloride, with production of acidity. Some vegetableresidues, only partly humified, produce the same result.Incidentally, the results, if correctly interpreted, furnish con-clusive evidence t h a t humic acids are not merely adsorbing colloids,but that they are chemically acid.Different manures affect soil acidity, due t o the presence of ironand aluminium salts, in different ways, chlorides, nitrates, andsulphates increasing the acidity, whilst, applications of acid phos-H.Kappen, Landw. Versuchs-Stat., 1916, 88, 13 ; 89, 39 ; A., i, 876AGRICULTUHAL CHEMISTRY AND VEGETABLE PHYSIOLOGY. 225phate, owing to the production of insoluble iron and aluminiumcompounds, diminishes the acidity.lTThe strongest argument in favour of adsorption, and against theexistence of true acids, in cases of soil acidity has perhaps beenthe failure t o show that such soils take up chemically equivalentamounts of different bases. It is now, however, pointed out thatthis law only holds good when all the resulting substances are ina true solution or when the solubilities of the partly soluble com-pounds which may be formed are of the same order.lB The possi-bility of secondary reactions has, moreover, t o be eliminated.Thedifficulties have been overcome by employing only small amountsof finely powdered soil and large volumes of salt solutions, andthe time was limited t o about a minute. Under these conditions,nearly equivalent amounts of different bases are used.Whilst some substances may be adsorbed more than others, aselective adsorption of ions from the common alkalis arid alkalineearths seems to be doubtful. I n any case, the amounts concernedare only small, and might be atkributed t o chemical reactions withimpurities in the adsorbent. The acidity of cotton washed withwater, which was attributed by Cameron to selective adsorption, isnow thought to be due t o the presence of insoluble fatty acids.The acidity observed in soils, whether active or latent, is muchgreater in extent, and is in every way comparable with chemicalaction between acids and bases, etc.In upland soils, most of the acids causing acidity are insoluble,and the more the soil is leached with water the 11201'0 acid itbecomes, owing to the washing out of the bases liberated byweathering.Sodium and calcium are removed to a greater extentthan potassium and magnesium. When alkali and alkaline earthbases are not present in sufficient amounts, the acid silicates maytake up iron and aluminium, as is the case in acid soils.It has been pointed out that the acidity of soils, as hithertoestimated, is the total, or potential, acidity, whilst it is probablethat the effects of the acidity depend rather on the intensity ofthe acidity, which these methods fail to show, than on the amountof acid.lg The intensity of acidity, or of alkalinity, has beenestimated by measuring the hydrogen-ion concentration.Thehydrogen-ion exponents of a number of soils which were investi-gated were found t o vary between 4.4 and 8.6, an exponent of7 indicating neutrality, whilst the lower and higher number.,indicate respectively acidity and alkalinity.l7 S. D. Conner, J . I n d . Eng. Chern., 1916,8, 35 ; A . , i, 350.l9 L. J. Gillespie, J . Washington Acad. Sci., 1916, 6, 7 ; L4., i, 303.E. Truog, J .Physical Chem., 1916,20, 457 ; A . , i, 591.REP.-VOL. XIIT. 226 ANNUAL REPORI'S ON THE PROGRESS OF CHEMISTRYI n all these experiments, soil suspensions, and not extracts, wereemployed. I n acid soil extracts difficulties occur, owing to theslow reduction of nitrates, which, however, has no appreciableeffect when suspensions of soil are employed, owing probably tothe large excess of potential acidity.20 The amount of wateremployed in relation to the soil seems, within wide limits, to havevery little effect on the results, so t h a t it may be assumed t h a tthe ion concentration of soil suspensions and soil solutions areapproximately the same. The presence of carbon dioxide does notseem t o have any appreciable effect in acid soils; in alkaline soils,on the other hand, the reaction depends largely on the equil-ibrium between gaseous carbon dioxide, dissolved carbon dioxide,C03-ion, and HC0,-ion.Addition of potassium, sodium, and barium chlorides to soilsuspensions resulted in a distinct increase in the hydrogen-ion con-centration, which, on the whole, was greater when barium chloridewas used.All the results seem to point to the presence of soluble acids asthe cause of soil acidity and to make it unnecessary t o supposet h a t physical adsorption has anything to do with it.Althoughthere may be nothing improbable in a limited adsorption of a com-pound, the theory of selective ion adsorption seems, among otherthings, to disregard the ionic equilibrium.Some experiments in which a number of soils were treated withN-solutions of potassium nitrate 21 showed that whilst calcium andmagnesium were always dissolved, the more acid extracts containedaluminium as well, but no iron.The conclusion is drawn t h a t theacidity of extracts is caused by a hydrolysed calcium salt, and thatthere is no excess of either acid or base unless originally presentin the soil.The Organic Matter of Soils.An investigation in Sweden of mineral soils containing largeamounts of humus showed t h a t mechanical analyses are superfluousfor the purpose of classification, and t h a t the best indications areobtained by estimating the weight per litre and t h e maximumamounts of water which can be mixed with the soils without causingthem to flow.22The results of earlier investigations on soluble humus, carriedout for many years in California, indicated t h a t whilst arid soilscontain relatively small amounts of humus as compared with humidsoils, the percentage of soluble nitrogen is far higher in the former2 0 L.T. Sharp and D. R. Hoagland, J . Agric. Research, 1916,7, 123.21 F. E. Rice, J . Physical Chem., 1916, 20, 214 ; il., i, 360.22 A Atterberg and S. Johansson, Internat. Mitt. Bedenk., 1916,6, 38AGRICULTURAL CHEMISTRY AND VEGETABLE PHYSIOLOGY. 227thaii in the latter, so that the total soluble nitrogen in the twotypes of soils tends to be equalised. To some extent this view issupported by the bacteriological results already referred t0?3 whichshowed that the organisms which destroy cellulose are more activein arid than in humid soils, whilst the organisms which directlyaffect the nitrogen content of the soils, whether by increasing orretarding losses of nitrogen, seem to be more active in humid soils.No method of estimating humus has, however, as yet been devisedwhich can be considered quite satisfactory, and the evidence nowobtained seems to show that by extracting soils with ammoniathe results may be far too high, even when the extract is boiledfor some hours with magnesia before estimating the nitrogen.2JIt is suggested that in soils from humid regions, the error wouldbe diminished by the greater absorptive power of such soils forammonia, whilst with arid soils, which are less absorbent, the errorwould be increased.Whatever the cause of the high resultspreviously obtained with arid-soil humus, recent analyses, in whichthe humus was extracted with sodium or potassium hydroxide, didnot give nit,rogen percentages higher than 7 or 8 per cent., andindicate that the humus of arid soils is not more nitrogenous thanthat of humid soils.The observations of Hilgard and Jaffa showed that when soilsare thoroughly extracted with ammonia, no coloured substance isobtained by extracting with sodium or potassium hydroxide, andvice versa. The extracts are, however, not necessarily identical,since the one solvent may dissolve more non-coloured matters thanthe other.25A number of estimations in which soiIs were extracted with4 per cent. potassium hydroxide for nine days, which is shown togive the maximum amount, of humus, gave results agreeing fairlywell with those obtained by the Hilgard-Jaffa method.Five out ofsixteen arid soils were found t o contain humus with more than10 per cent. of nitrogen, whilst 110 humid or semi-arid soils gavesuch high results.The recent results, whilst somewhat conflicting, would seem, a tany rate, t o show that high percentages of nitrogen in arid-soilhumus are not sufficiently general to enable a distinction betweenarid and humus soils to be made on the strength of the characterof the humus as regards its nitrogen content.The results of an investigation of Nebraska loess soils showed26a3 C. B. Lipman, Zoc. cit.24 C. B. Lipman, Soil Sci., 1916,1,283.25 F.J. Alway and E. S. Bishop, J. Agric. Research, 1916, 5, 909.26 F. J. Alway and M. J. Blish, Soil Sci., 1916, 1, 239.1 228 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.that the aniount of humus decreases from east to west, beingcoincident with a decrease in the rainfall from more than 750 mm.to less than 500 mm. The percentages of nitrogen in the humusfrom the humid and the semi-arid districts do not, however, showany marked differences. The soluble pigment in the surface soildecreases with the humus, whilst with the subsoil there was anincrease from the third to the sixth foot. Comparison of thecolour of subsoils was found t o be difficult, owing to colouringmatters other than the soluble pigment, and the colour of the sub-soil and the amount of the soluble pigment have n o definite rela-tion to the total nitrogen and the soluble humus.The soils, even in the most easterly portion of the transitionregion, seem to be paler in colour and shallower than the typicalRussian black soils, and in the west resemble rather the brownsoils which occur near the borders of the black soil zones in Russia.With reference t o the pigment,27 it has been shown t h a t it prob-ably rarely exceeds 40 per cent.of the humus, and that it con-tains relatively low amounts of nitrogen, and is therefore of nogreat importance in connexion with soil fertility. It was foundin all mineral soils, but not in acid peats.Apart from the pigment, the forms of nitrogen in soil humusdo not seem to differ much from those of non-humified vegetablesubstances, except t h a t the latter contain large amounts ofnitrogen soluble in 1 per cent. hydrochloric acid.As alreadysuggested, ammonia and sodium hydroxide in 4 per cent. solutionsextract different substances as well as different amounts from soils;it was found t h a t the former extracts less carbon and more colourthan the latter.To the now fairly long list of definite organic corr!pounds whichhave been found to be associated with soil humus, a-crotonic acidhas recently been added.28 The substance was isolated from afine, sandy loam in Texas, which is deficient in lime and is in-sufficiently drained; the soil has a high reducing power and asomewhat low oxidising power, so that the conditions are favour-able for the production of organie acids.The presence of this acidmay be due to the hydrolysis of vinylacetonitrile (ally1 cyanide)which occurs in some plants, or it might be produced fromP-hydroxy-acids formed during the destruction of cellulose.Satisfactory evidence has been obtained that the phosphorusdissolved from soils by dilute alkali is present largely in organicforms,29 so t h a t the view of Grandeau, who held that the phos-27 R. A. Gortner, Soil Sci., 1916, 2, 395,2 8 E. H. Walters and L. E. Wise, J . Agric. Research, 1916, 6, 1043.z9 R. S. Potter and T. H. Benton, Soil S c i . , 1916,2, 291AGRICULTURAL CHEMISTRY AND VEGETABLE PHYSIOLOGY. 229phorus is entirely organic, is more nearly correct than t h a t ofvan Bemmelen, who believed it to be absorbed inorganic phos-phorus.Very little is known about' the organic phosphorus com-pounds present in the organic materials usually applied to soils,and in addition t o the nucleoprotein and other substances thusadded, the production of organic phosphorus compounds frominorganic materials by bacteria may be of some importance. Itwas found t h a t soils which received the more inert manures, suchas peat and straw, contained less organic phosphorus, in relationto the total amount, than other soils. It will be desirable t o studythe behaviour of plants towards organic phosphorus compounds,very little, if anything, having been done in this direction beyonda few experiments with lecithins.Soil Toxim.The results of an investigation on the toxic action of decayedvegetable substances30 showed t h a t the products of the decay ofpotatoes, turnips, and ilrymphaea rhizomes are all toxic to plants;i t was also found that fresh Nymphnen rhizomes, when extractedwith water under 33 kilos.pressure, yield a substance injurious togrowing plants.From decayed Nymplznea rhizomes, toxic substances wereobtained by directly extracting with water. The extract retainsits toxic properties when extracted with ether, and the etherextract is also toxic. When the decayed rhizomes are distilledwith water a t 40°, both the distillate and the residue are toxic.The toxicity of the solutioiis when diluted is inore or lessdestroyed by neutralising with sodium hydroxide, and it is removedby animal charcoal, whilst the extract, even when not diluted, isrendered harmless by precipitation with ammonium sulphate.Aeration, obtained by frequently shaking 50 C.C.of the solutionin a large flask for three days, failed t o diminish its toxicity.The conclusion drawn from these results is that a t least threeclasses of somewhat toxic substances are formed or released duringthe decay of many vegetable matters-colloids, very volatile sub-stances, and certain bases. Although these products are not readilyoxidised in solution, it' seems probable t h a t when formed in soilsthey would be more o r less rapidly destroyed or otherwise renderedharmless.Whilst the toxicity of extracts of non-productive soils can beovercome by treatment with charcoal and some other substances,it has been shown by field experiments, extending over a numberof years, that addition of carbon black t o the soil is without30 G.B. Rigg, Bot. Gaz., 1916, 61, 295230 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.effect.31 A number of experiments have now been made in whichporous vessels filled with carbon black were buried in the soil, sothat the substances absorbed by the carbon were removed beyondthe reach of thk roots. For soils in pots and on greenhousebenches, ordinary battery cylinders were employed, whilst in plotexperiments, dyain-pipes, specially made of thin cement, were used.The results of the pot and greenhouse experiments showed thatthe presence of carbon black, enclosed in this manner, increasedthe yields of plants grown in a poor, unproductive soil and in soilto which salicylic acid and vanillin respectively had been added.I n the field experiments it was shown t h a t by burying poroustubes filled with carbon black, charcoal, and calcium carbonate inthe soil, the yield of cowpeas was considerably increased, especiallyon the plots which had charcoal.The amounts of substances absorbed were too small for identifi-cation.It was found, however, that, all three absorbents containeda small amount of fatty acids, and t h a t the calcium carbonatecontained aldehydes as well. The latter were probably absorbedby the carbon and then oxidised.Soil F ~ ~ n g i and Bacteria.I n a preliminary communication from the U.S.Bureau of PlantIndustry, containing the results of an investigation of twenty-four Azotobacter cultures and eighteen strains of otherorganisms,32 it is shown t h a t all the bacteria studied exist altern-ately in an organised and in an amorphous stage, in which thecontents of the separate cells become mixed through the dis-integration of the cell walls. From this “symplasm” new cellsare formed, which develop in different ways, according t o thedifferent formation and quality of the symplasm. Finally, cellsof normal type are again produced.The wide morphological differences obtained with A zotobacteroccur under suitable conditions with all bacteria, and the develop.ment of the bacteria is characterised by the regular occurrenceof many different forms definitely related to each other.I naddition to the formation of the symplasm, it is found t h a t thecontents of the cells may interact by the direct union of two ormore individual cells.Apart from the value which, if confirmed, the results shouldhave in connexion with systematic bacteriology, it is suggestedt h a t they afford an explanation of the regular seasonal variationsin the activity of soil organisms, which may be due to the effect31 J. J. Skinncr and J. H. Reattic, Soil Sci., 1916, 2, 93.32 F . Lijhnis and N. R. Smith, J . Agric. Research, 1916, 6, 675AGRICULTURAL CHEMISTRY AND VEGETABLE PHYSIOLOGY. 231of season on the different ways of multiplication and propagationof the bacteria.I n connexion with the sterilisation of soils, it has been foundthat the intermittent heating of moist soil in air at 8 2 O for fivedays is very effective, whilst the soluble matter of the soil isincreased 46 per cent.as compared with sixteen times that amountwith steam sterilisation.33 Indications were obtained that anintermittent application of volatile antiseptics, in a partialvacuum or in conjunction with heat, for more than three dayswould result in complete sterilisation without any materialchemical change in the soil.A good deal of attention has been given to the fungi of soils,which have been somewhat neglected until recently, and therecan be no doubt that these organisms are of considerable importiance as agents in the production of food for the higher plants.The examination of a number of soils revealed the presence ofmore than a hundred distinct species of fungi, belonging t o thirty-one genera.34 Indications were obtained that a distinct fungusflora exist?, but this can only be established by the investigationof many soils from different localities.Many of the fungi were found to have the power of decom-posing cellulose, and their activity would seem chiefly to beexercised in the production of humus and in the liberation ofammonia from organic compounds. The greatest accumulationof ammonia was found to take place during the period precedingspore formation.35Experiments in which sterilised soil was inoculated with threedifferent fungi 36 showed that' ammonification of cottonseed mealand of dried blood only takes place within comparatively narrowlimits of reaction, and that it is greatest between the neutralpoint and an acidity corresponding with 2240 kilos.of lims perhectare; greater acidity o r alkalinity, however slight, are un-favourable. From these results, i t seems likely that fungi are.important for ammonification when the conditions are unf avour-able for bacteria.As regards the influence of different forms or organic matteron the activity of fungi, i t has been found that substances ofvegetable origin are, on the whole, the most suitable.37 Theresults vary, however, a good deal with closely allied groups oforganisms supplied with the same organic matter. The behaviour33 D. A. Coleman, H. C. Lint, and N. Kopeloff, Soil S c i ., 1916, 1, 259.34 S. A. Waksman, ibid., 1916, 2, 103.35 S. A. Waksman and R,. C. Cook, ibid., 1, 275.36 N. Kopeloff, ibid., 541. 3' D. A. Coleman, ibid., 2, 1232 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.of the different fungi in relation to aeration and towards manuresshowed considerable variations.A fungus has been found in soils, stra-8, and in animals whichdigests cellulose and readily utilises mannitol and ethyl alcoho1.38I n the latter substance it is remarkably resistant; it will growin 80 per cent. alcohol, whilst even 96 per cent. alcohol does notdestroy the spores. When a culture solution containing celluloseis filtered through a Berkefeld filter, the filtrate, free from cells,continues to dissolve cellulose for several weeks.The group of Actkomyces, which form an important portion,as regards numbers, of the soil flora, has been found to occurmuch more abundantly in grass soils than in arable soil.39 Asactive agents in the decomposition of humus, their chief import-ance is probably connected with the formation of humus.I n theliberation of ammonia they seem to be of minor importance.40I n a more general investigation of the organisms which decom-pose cellulose, it was found that sixty-nine Californian soils allcontained one o r two active forms, and t h a t most of the specieswere found in soils from widely separated districts.41 Fifteennew species were isolated, in addition to seven others. Theorganisms assimilate the nitrogen of ammonium salts, nitrates, andorganic compounds, peptone being the most suitable of these andcasein the least suitable.The importance of cellulose bacteria in soils is evident, as untilthe cellulose is decomposed it is useless t o other organisms. Onthe other hand, the value of cellulose would be much less if itwere a t once converted into carbon dioxide and methane, whichoccurs under some conditions. In these experiments, no evolutionof gas was observed, and it seems possible that, under naturalconditions, the combined efforts of different organisms may resultin the production of nitrogenous humus from cellulose and freeor inorganic nitrogen.I n a promising investigation on the activities of soil bacteriain relation to the yield of crops when maize or clover was growncontinuously, and in different rotations, of two to four years,indications were obtained t h a t certain activities-ammonificationand nitrification-are closely associated with crop yields.45 Fix-ation of nitrogen seeiiis to be less trustworthy as an indication offertility, and can only be depended on in certain cases.I t was38 W. Ellonberger, A. Scheunert, W. Grimmer, and A. Hopffe, Zeitsch.physiol. Chem., 1915, 96, 236 ; A., 1916, i, 588.39 E. J. Conn, N e w YorkExper. Stat. Techn. Bzdl., 52, 1016.40 S. A. Waksman, and R. E. Curtis, Soil Sci., 1916, 1, 90.41 J. G. McBeth, ibid., 437.45 P, E. Brown, J. -4gric. Research, 1916, 5, 865AGRICULTURAI, CHEMISTRY AND VEGETABLE PHYSIOLOGY. 233found, for instance, that greater crops were obtained in a three-year rotation than in a two-year rotation, and t h a t the latter gavehigher results than the continuous growth of the same crop.Thebacteriological tests usually showed exactly the same relations.The experiments, which were continued for a second and thirdyear, showed t h a t the plots ranked differently each year, owingchiefly to altered seasonal conditions, both as regards crop yieldsand, coincidently, the bacterial activity of the soils, so that,independently of season and with varying amounts of soil mois-ture, crop production and the activity of the bacteria remainedrelatively similar.As regards the effect of farmyard manure, applications up t o16 tons per acre resulted in increased yields, whilst with 20 tonsthe yieId was less than with 1 2 tons.Both ammonification andnitrification experiments agreed with the crop results, bacterialactivity being increased by farmyard manure up to 16 tons andretarded by larger amounts. The crop yields and the bacterialactivity, including fixation of nitrogen, were also increased by2 tons of ground limestone; with larger amounts, the bact,erialactivity alone was increased.The effects of continuous cropping and a rotation on soil bacteriahas also been investigated on plots which have received the sametreatment for twenty-five years.46 It was found that, on the un-manured plots, the numbers of bacteria were greatest wheretimothy was grown continuously, next on the rotafion plots, andleast on the continuous maize and wheat plots.Addition of farm-yard manure considerably increased the number of bacteria, and,at the same time, almost reversed the order of the plots, the maizeand wheat plots having much the greatest numbers of bacteria,and the timothy and rotation plots the lowest. The ammonifyingpower of the soil seemed t o have no relation t o the number ofbacteria and to' remain practically unaltered under the differentsystems of cropping. The nitrifying power of the soils was foundt o vary considerably, the continuous growth of maize and wheat,,without manure, resulting in a relatively low nitrifying power,whilst addition of farmyard manure, and, to a less extent, ofchemical manures, increased the nitrifying power of the soils,especially on the plots under continuous maize and wheat.I n another series of experiments, made in pots, different amountsof farmyard manure and water were added to a calcareous soil,which was then left for four months.47 Inoculation experimentswith synthetic media then showed t h a t the greatest number of*G P.L. Gainey and W. E. Gibbs, J. Agric. Research, 1916, 6, 953.47 J. R. Greaves and E. G. Carter, ibid.,889.I 234 ANNUAL REPORTS ON THE PROGRESS O F CHEMISTRY.organisms developed when soils containing the largest amount offarmyard manure (25 tons per acre) were employed; the ammonify-ing and nitrifying powers of the bacteria were also highest inthese soils, whilst fixation of nitrogen was most active in the caseof the soils which had only 10 tons of manure.I n a fallow field soil, the bacterial numbers were found t o beincreased by farmyard manure up to 15 tons (the largest amountemployed), and by irrigation water up to 50 em.; ammonificationwas also increased by the same amount of manure and by 25 cm.of water, larger amounts of water having a depressing effect,especially in conjunction with farmyard manure.Nitrification wasincreased by farmyard manure and diminished by irrigation.As compared with fallow, cropped soil showed lower bacterialnumbers and lower animonifying and nitrifying powers. I naccordance with results previously referred to, a close relationwas found t o exist between the crop yields and the bacterialnumbers and bacterial activity on soil with 5 and 15 tonsrespectively of farmyard manure.Under ordinary conditions, the nitrogen-carbon ratio in soilstends to become narrower as time goes on, until only the moreinert substances remain, so that when this ratio is found t o be anarrow one, it is likely that the soil is deficient in fresh organicmatter.48Experiments in which a considerable variety oE organic manureswas added to a soil showed that, all the common humus-formingmaterials were favourable to increased bacterial activity, and thatthe effects produced depended rather on the chemical constitz-tion of the substances added than on their iiitrogen-carbon ratios.As regards ammonification, the greatest effects were produced byhorse and cow manures, rotted manure and timothy hay, whilststraw had less effect and leguminous hays least of all.Nitrifica-tion was influenced much in the same manner, except that legu-minous green manures gave higher results than the non-legumes.For fixation of nitrogen, straw and non-leguminous hays werealmost as efficient as dung, whilst leguminous hays gave the lowest.results. It may, accordingly, be more economical to employ non-leguminous straws, and thus increase fixation of nitrogen, than toadd the more nitrogenous and more costly materials, which areless favourable to nitrogen fixation.Manures with narrow nitrogen-carbon ratios increased the yieldof oats; those with wide ratios decreased the yields, the nitrogenof the manure being of importance in the soil employed. Thesecond crops were, however, benefited as much by non-leguminous4 8 P.12. I3rou.n and F. E. Allison, Soil S c i . , 3 910, 1, 40AGRICULTURAL CHEMISTRY AND VEGETABLE PHYSIOLOGY. 235as by leguminous plants, owing to increased fixation of nitrogenin the longer interval.It has been found t h a t addition of dried blood t o a sandy loam49resulted in a rapid production of ammonia, which continued foreighty-six days, after which there was a marked decrease untilthe end of the experiment, which lasted 240 days. I n the wholeperiod, 79 per cent. of protein nitrogen was converted intoammonia, whilst of the monoamino-acids about 89 per cent. wasammonified. Arginine and histidine each gave up about 83 percent. of their nitrogen. The soil was found t o contain substanceslike protein, soluble in 1 per cent.sodium hydroxide; it is, how-ever, doubtful whether these are residues from the dried bloodor whether they are newly formed proteins.The results of experiments with a large number of humid andarid soils showed t h a t the latter do not, as Hilgard supposed,nitrify more intensely than the former.50 Ammonium sulphateand cotton-seed meal were nitrified more vigorously in the aridsoils, whilst with dried blood and with the soil nitrogen itself,nitrification was more active in the humid soils. Dried blood isammonified very rapidly, and in soils with relatively low adsorp-tive power the ammonia may become strongly toxic to nitrifyingorganisms. The dissolved organic matter in arid soils may alsohave an inhibiting effect.A number of experiments have been made on the influence ofvarious salts on the aminonification of dried blood in a silt l o a d 1Taking first the salts of sodium, potassium, calcium, and mag-nesium, of which the chlorides, sulphates, nitrates, and carbonateswere employed, it was found that sodium sulphate, potassiumchloride and sulphate, calcium chloride and nitrate, and mag-nesium nitrate all failed to stimulate ammonification in any con-centration, whilst all the other salts increased ammonificationin some concentration.As regards toxicity, it is shown t h a t usually chlorides are themost injurious, then nitrates, sulphates, and carbonates, and t h a ttoxicity depends more largely on the electronegative than on thepositive ions.Compounds which are toxic in the lowest concen-tration when different salts are compared are not necessarily themost toxic in higher concentrations, the toxicity of some saltsincreasing with concentration more than t h a t of others, so thata salt which in low concentrations acts as a stimulant may, asthe amount is increased, be more toxic than a corresponding49 E.Lathrop, Soil Sci., 1916, 1, 509 ; A . , i, 703.5 0 C. B. Lipman, P. S. Burgess, and M. A. Klein, J . Agric. Research, 191'3,7, 47. 51 J. E. Greaves, Soil Sci., 1916, 2, 443.I* 236 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.amount of a salt which has no stimulating action in any con-centration.Iron salts in small amounts stimulated amrnonification, thegreatest effect being obtained with the chloride (5.8 per million ofiron), whilst with the other salts twice as much had to be addedto produce the maximum stimulation, which was much less markedwith the sulphate and the chloride.With increasing concentra-tion, the toxic point was observed first with ferric sulphate, whilstin the higher concentrations the chloride was most' toxic.I n the case of manganese it is shown that, the sulphate, whenapplied a t the rate of 17.2 per million of manganese, increasedthe production of ammonia 23.8 per cent., and t h a t the nitratealso gave a considerable increase. The carbonate, which is usuallysupposed t o have the greatest% stimulating effect on higher plants,had much less influence on amrnonification, whilst it seems doubtfulwhether the chloride can act a t all as a stimulant.The amounts of sodium, calcium, potassium, and magnesiumwhich reduce ammonificatioii to half normal have the same effecton the growth of wheat; the other salts seem to be less toxic withammonifying organisms than with higher plants.I n another series of experiments on the ammonification of driedblood, this time in a clay loam, it was found that manganesechloride is toxic in amounts of 0.03 per cent.or more, whilst withsmaller amounts ammonification was perhaps slightly stimulated.52Manganese sulphate had a decidedly stimulating effect whenapplied a t the rate of 0.005 t o 0.09 per cent.; with the nitrate, theresults with the smallest amounts, although somewhat conflicting,seem to point to a slight stimulation. With 0.1 per cent.ofmanganese oxide, ammonification was distinctly retarded ; largeramounts failed to produce any further decrease.I n nitrification experiments with the same clay loam, it wasfound t h a t small amounts of manganese chloride have a distinctlystimulating action ; the smallest amounts of sulphate stimulatednitrification, whilst a slightly increased amount, sufficient t o retardammonification, had no effect either way on nitrification, whichwas also retarded by the nitrate and oxide.Applications of 0.5 t o 1 per cent. of sulphur have been foundto result in a great increase in the number of soil organisms; theaction is slow, and no reduction in numbers was observed untilafter forty-four days.53 The same amounts of sulphur increasedammonification considerably, whilst nitrification was decreasedafter thirty days, owing probably to the increasing acidity.52 P.E. Brow-n and G. A. Minges, Xoil Sci., 191G, 2. (37.63 W. Pitz. J. Aqric. Research, 1916, 5, $71 ; A., i, SSOAGRICULTURAL CHEMISTRY AND VEGETABLE PHYSIOLOGY. 237Sulphoficcction.Experiments made in a loam well supplied with organic mattershowed that monocalcium phosphate, acid phosphate, rock phos-phate, and small amounts of calcium sulphate all increase thesulphofication of free sulphur in the ~0il.54 Calcium carbonate,applied in the usual amounts t o acid soils, considerably increasedthe oxidation of sulphur, and even very large amounts actedf avourably. Magnesium carbonate increased sulphofication onlywhen small amounts were applied.I t has been suggested that the formation of sulphuric acid fromsulphur by means of soil organisms might be utilised for the pro-duction of available phosphoric acid from insoluble phosphaticmanures, and a number of experiments have been made in whichsulphur and rock phosphate were mixed with sand and with twodifferent soils, all of which were inoculated with a soil infusi0n.5~The results are promising, especially those obtained with a redloam, in which 1.982 grams of soluble phosphoric acid (an increaseof 1,840 grams) was produced in 100 grams of soil in fifteenweeks.In another loam the gain was less than half that amount,whilst in sand it amounted t o 0.282 gram.The production of acid in sand was greater when both sulphurand phosphate were added than with sulphur alone, and the maxi-mum acidity was reached in the tenth week.I n the red soil, withsulphur only, the acidity increased up t o the end of the fifteenthweek, although only slowly after the tenth week. Whilst thesoluble phosphoric acid in sand showed a relatively slight increaseafter the tenth week, the most striking gain in the red loam wasfrom the twelfth to the fifteenth weekThe results seem to justify the conclusion that i t will be practic-able to produce soluble phosphoric acid by adding insoluble mineraIphosphates to compost heaps in which sulphofication is active.Fixation of f l i t r o g e ? ~ .In continuation of experiments on the action of arsenic com-pounds on soil organisms, which showed that both ammonificationand nitrification are stimulated by small amounts of these com-pounds, the behaviour of nitrogen-fixing organisms has now beeninvestigated.56The compounds employed in t,hese experiments were sodium and54 P.E. Brown and H. W. Johnson, Soil Sci., 1916, 1, 339.B5 J. G. Lipman, H. C. McLean, and H. C. Lint, ibid., 533.56 J. E. Greaves, J . Agric. Research, 1916,6, 389235 ANNUAL HEPOR1'S ON THE PROGHESS OF CHEMISTRY.lead arsenates, zinc arsenite, arsenic trisulphide, and Paris-green,and i t was found that all these substances, except, the last,stimulate nitrogen fixation in soils. The most' active stimulant islead arsenate, which does not become toxic until the amount ofarsenic exceeds 400 per million.Zinc arsenite is the least activestimulant, whilst Paris-green was found t o be very toxic, owing t othe copper and not t o the arsenic it contains.The stimulating action of arsenic compounds, which is greatestwhen the water-soluble arsenic amounts t o about 10 per cent., wasobserved in soils which varied considerably in chemical and physicalproperties.zotobncter showed that onlyone of them was stimulatled by arsenic; with this type the increasein the amount of nitrogen fixed was considerable.The increased fixation of nitrogen, resulting from the additionof compounds of arsenic, is attributed to the more economicalutilisation of carbonaceous food, t o the destruction of undesirablesoil organisms, to thel production of soluble phosphorus, and t o astimulating action on the cellulose organisms, resulting in a morerapid production of carbonaceous materials suitable for consump-tion by the nitrogen-fixing organisms.Comparing the behaviour of ammonifying, nitrifying, and nitro-gen-fixing organisms towards arselnic compounds, it is shown that400 parts of arsenic as lead arsenate has practically the same effecton the last two, whilst the ammonifying organisms are retardedmore ; with arsenic trisulphide the effect on the three organismsis about the same, and with the same amount of arsenic as zincarsenite the nitrifying and nitrogen-fixing organisms act normally,whilst the activity of ammonifying cwganisms is considerablydepressed.With lead arsenate and with arsenic trisulphide greater amountsare required, for maximum stimulation, by nitrogen-fixing organ-isms than by the others.Zinc arsenite, on the other hand, has t obe supplied much more liberally to nitrifying organisms than t othe other two. It! must be borne in mind that in these coin-pounds the sulphur and the zinc can act as stimulants in additiont o the arsenic.The results are of considerable interest, and make i t evidentthat in soil culture experiments with higher plants the results,favourable! o r otherwise, obtained when small amounts of varioussubstances are added t o the soil may be due, not to a direct actionon the plants, but to increased fertility, or infertility, due to thestimulation of some organisms and the suppression of others.Experiments with three types oAGRICULTURAL CHEMISTRY AND VEGETABLE PHYSIOLOGY.239Chenzistry of i h e Growing Plant.An investigation of the leaves of mangolds has shown t h a tstarch is almost entirely absent from the time when the root beginsto develop, and the sugars begin to migrate to it.57 Maltose is notfound a t all in the leaves and stems. I n the early stages ofgrowth the leaves contain an excess of sucrose over the hexoses,whilst, later on, when the sugar is being stored in the roots, theliexoses predominate in the leaf.I n the mid-ribs and stems there is always an excess of hexosesover sucrose; and whilst the amount of hexoses varies widelyduring day and night, and throughout tlie season, the sucroseremains practically constant.A steady and rapid increase in theratio of hexoses to sucrose was observed from leaves to mid-ribs,and so on to the bottom of the stalks.I n the leaf the proportion of sucrose closely follows the tempera-ture curve in the day time, whilst the hexoses increase faster, sothat the curve showing tlie ratio of hexoses to sucrose is practi-cally parallel to the temperature curve.The results accord with Brown and Morris’s view that sucroseis the primary sugar formed. I n the veins it’ is transformed intohexoses for the purpose of transportation ; and the proportion ofliesoses shows a progressive increase a s the root is approached. I nthe root the hexoses are converted into sucrose, which remains un-changed until required for the second year’s growth.Dextrose and laevulose are found in approximately equal amountsin the leavea of mangolds, whilst the results indicated an excess ofdextrose.58 It seems probable t h a t the two substances are reallypresent in equal amounts, as they would be if formed by inver-sion, and that the divergent results are due to optically active acidamides and amino-acids.I n potato leaves sucrose is the principal sugar when the tubersbegin to develop, whilst tlie amount of hexoses is quite small-1 per cent.or le~s.5~ From sunrise to 2 p.m. the amount of sucroseincreases, and then falls during the rest of the day and night;the hexoses fluctuate considerably during the early part of the day,owing, perhaps, to conversion into starch, or to production fromstarch, the proportion of which changes very little up to 2 p.m.As soon as the sucrose reaches its maximum, a t 2 p.m., solublestarch is first detected, and it increases in quantity up to 6 p.m.;coincidently, the hexoses also increase, as a result, probably, of57 W.A. Davis, A. J. Daish, and G. C. Sawyer, J . Agric. Sci., 1916, 7, 256.5 8 W. A. Davis, i b i d . , 327.5 9 W. A. Davis and G. (2, gawyer, ibid., 352240 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.the hydrolysis of the sucrose. The starch reaches its maximum a t6 p.m., after which both the starch and the soluble starch decreaserapidly until midnight to 2 a.m., when only 0.2 per cent. remains.I n t h e stems the reducing sugars predominate considerably overthe sucrose, owing, probably, to the inversion of the sucrose for thepurpose of translocation.The same difficulty in estimating the proportions of dextrose1 andlzvulose, as already referred t o in the case of mangolds, occurswith potatoes, so t h a t it cannot definitely be stated whether theyare present in equal amounts or not.It' is possible that the stemscontain an excess of dextrose, as the stems seem to contain lessoptically active impurities than the leaves.The degradation of the starch is probably brought about by amixture of enzymes, similar to takadiastase, and the series ofchanges is : starch, dextrins, maltose, dextrose.It has been shown that maltase is probably always present inplants in which starch degradation occurs.60 Germinated and un-germinated cereals contain considerable amounts, and it may bepresent in malt when the temperature of the kilning has been toolow to destroy i t ; and its presence in malt would account for theproduction of dextrose from starchThe maltase of plants does not act directly on starch or dextrins,and no direct cleavage of dextrose from starch ever occurs.Maltoseis first formed by ordinary diastatic enzymes, and is then attackedby maltase.Experiments with the leaves of a number of plants (Tropeolzcnz,potato, dahlia, turnip, sunflower, and mangold) showed t h a t thecrushed pulp acts on soluble starch with production of reducingsugars, chiefly dextrose with some maltose; so t h a t it seems evidentt h a t all these leaves contain maltase.61Evidence has been obtained of an indirect production in plantsof starch from dextrose.GZ Maize plants were grown in waterculturels, under sterile conditions, in presence of dextrose andmineral nutrients, but in absence of carbon dioxide.When theleaves were exposed to light, but not otherwise, starch was foundto be present. Thel presence of oxygen is necessary, and theportion of the solar spectrum which has most influence is thatwhich exerts the maximum effect in the formation of chlorophyll.The conclusion drawn from these results is t h a t the sugarabsorbed by t h e roots is oxidised in the plant to carbon dioxide,and t h a t the latter is converted into starch by the usual process.ibid., 56 ; A . , i, 535.6 0 W. A. Davis, Biochem.J . , 1916, 10, 31 ; A . , i, 535.61 A. J. Daish, ibid., 49 ; A . , i, 535.62 C . Ravenna, Atti R. Accad. Lincei, 1916, [v], 25, i, G49 : L4., i. 688.Also A. J. DaishAGRICULTURAL CHEMISTRY AND VEGETABLE PHYSIOLOGY. 241A somewhat similar experiment has been made with sugar beetwhich was grown on absence of carbon dioxide, but with variouscarbohydrates in addition t o nutrients, both with and withoutpot assiuin .63I n presence of light the plants produced proteins, both with andwithout the presence of potassium, when sucrose, dextrose, orlzvulose were added. I n absence of light, proteins were formedwhen nitrates were added along with the carbohydrate; the pres-ence of a potassium salt is, however, essential under these condi-tions.The importance of potassium in the synthesis of proteinscan only be shown in absence of light.A similarity has been shown to exist between the absorbingpower of plants for saline solutions and that of ~0ils.6~ When thestems of certain plants were immersed successively in distilledwater, a 1 per cent. solution of potassium, o r ammonium chloride,and again in distilled water, the presence1 of calcium was readilyd-etected in the cliloride solutions, but not in the distilled water.The base, potassium o r ammonium, which takes the place of theca.lcium originally present, can, in turn, be removed by immersingthe stem in a solution of calcium chloride. Chlorides, nitrates, sul-phates, and carbonates of potassium, sodium, ammonium, lithium,and magnesium all produce decalcification.Plant Nutrition and Stimulants.The quest'ion of the calcium-magnesium ratio, which is nowtwenty-five years old', continues to receive attention.A series ofexperinients? extending over three years, and including about threehundred pot cultures, indicated that the only thing that reallymatters is that sufficient amounts of both substances are present,and that the ratio, within wide limits, had no effect a t all.65Wheat, soja-beans, lucerne, and cowpeas all grew normally in96 per cent. of dolomite and 4 per cent. of sand, in magnesianlimestone, and in sand containing 7 per cent. of magnesite. Pre-pared magnesium carbonate applied t o a silt loam had no injuriouseffects up to 0.7 per cent., whilst in sand, 0.35 per cent.inhibitedgrowth.The percentages of calcium and magnesium in wheat straw,grown under different conditions, varied considerably, whilst theyields remained the same. Plants do not necessarily take up thetwo substances in the proportions in which they are applied.Incidentally, it was found that lucerne, grown in sand ex-J . Stoklasa, Biochem. Zeitsch., 1916, 73, 107 ; A . , i, 354.G4 H . Devaux, Compt. rend., 1916,162, 561 ; A., i, 457.G5 F. A. Wyatt, J. Agric. Research, 1916,6, 589242 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.tracted with concentrated hydrochloric acid, was found to containf a r more calcium and magnesium than the original seed con-tained.At Woburn, the application of ground magnesia to a soil, so asto alter the lime-magnesia ratio from 1.6 t o 1.1, increased the yieldof wheat grain and straw, and the1 percentage of nitrogen in thegrain .66I n pot experiments the yield of wheat, grown in a soil contain-ing 1.06 and 1-89 per cent.respectively of lime and magnesia, wasconsiderably increased by addition of calcium carbonate. I nanother soil, containing more lime than magnesia, applications ofmagnesia were beneficial up to the point a t which the calciumand magnesium became equal.Boric acid and borax have been shown to be toxic to wheat andbarley when the soil contains only 0*0005 per cent. of boron inthese forms.67 Half of that amount of boron had, however, aslightly stimulating effect.Another series of experiments with boron has been made inconnexion with the employment of borax and colemanite as larvi-cides.68 It was found that leguminous plants may be injured by0.0011 per cent.of boron, whilst other plants were not affected bythat amount. No stimulating action was observed, owing, perhaps,to the amounts used not being sufficiently small. Whilst all theplants were found to contain boron-even those grown in soilwithout added boron-the amounts taken up varied a good deal;there was a tendency for the boron to accumulate in differentparts of the plants, exceptl in the case of leguminous plants, inwhich it distributed itself more evenly.Application of sulphur to a silt loam has been found slightlyt o increase the growth of red clover, without, however, influencingthe formation of rootinodules.~g I n Ohio, both sulphur and hydro-gen sulphide were found t o increase the yield of red cl0ver.7~ I nanother series of experiments, applications of sulphur, whilst some-times beneficial, were more frequently injurious, owing, possibly,to incomplete oxidation to sulphites.71 I n a soil containing plentyof organic matter, sulphur failed to have any effect a t all.'?Comparatively few experiments have been made with strontium.At Woburn i t has been found that the chloride, supplied a t the66 J.A. Voelclrer, J. Roy. Ayric. SOC. Engl., 1915, 76, 331, 351.6' J. A. Voelcker, loc. cit., p. 347.6 8 F. C. Cook, J . Agric. Research, 1916, 5, 877 ; A . , i, 302.6 9 W. Pitz, Eoc. cit.7 0 J . W. Ames and Q. E . Boltz, Ohio Agric.Exper. Stat. Bull., 292, 1916.71 E. R. Hart and W. E. Tottingham, J . Agric. Research, 1915, 5, 233.72 T. Pfeiffer and W, Simmermacher, Bied. Zentr., 1916, 45, 18 ; A., i, 196AGRICULTURAL CHEMISTRY AND VEGETABLE PHYSIOLOGY. 243rate of 0.1 per cent. of strontium, is distinctly toxic, whilst thesulpliate gave negative results.73Aluminium, as chloride, has been found to be very toxic to riceseedlings, growth being inhibited by solutions stronger thanN / 7500.74Lead nitrate increased the growth of a number of plants, grownin water cultures, and in field experiments both sugar beet andwheat were benefited.75 I n the case of oats the yield of grain wasincreased, whilst the yield of straw was diminished.The examination of a considerable number of wheat samples,obtained from differentl countries, invariably showed the presenceof manganese, the amount being approximately the same as thatof iron.76Whilst applications of manganese compounds have frequentlybeen beneficial to crops, the number of negative results has beenlarge, and the most recent results are far from being encouraging.Vegetation experiments with different plants in water cultures,pot cultures, and field experiments all failed to show any increasedgrowth with manganese, whilst iron, and even copper, acted favour-ably on the growth of lupines.77I n another series of experiments, which did show some gainunder the influence of manganese sulpliate, i t was found thatsodium sulphate had a precisely similar effect.78Almost equally unsatisfactory results have been obtained in aseries of experiments extending over three years.79 Indicationswere obtained t h a t under favourable conditions more nitrogenmay be assimilated when manganese is applied, owing to anexchange of bases.As regards stimulation, the results are largelynegative.M a nu r es.Some experiments on the effect of the straw in farmyard manureshowed t h a t whilst a mixture of cow and horse manure, withoutstraw, lost rather more than 5 per cent. of its nitrogen in fourweeks, the addition of 8 per cent. of wheat straw resulted in gainsof nitrogen amounting to 3-7-4.8 per cent.80 I n each case therewas a great increase in the number of bacteria, especially whenstraw was added. L4zotobnctcr was not identified; it was, how-73 J . A. Voelcker, loc. c i t . , p. 344.iz K. Niyake. J. Biol. Chem., 1916,25, 23 ; A., i, 590.7s A. Btutzer, J . Landw., 1916, 64, 1 ; -1., i, 704.76 W. P. Headden, J . Agric. Research, 1915, 5, 340.7 7 H. Vageler, Landw. T7ersuchs-Stat., 1916, 88, 159 ; A., i, 457.7 8 G. Masoni, Staz. sper. agrar. ital., 1915, 4$: 822 ; A , , i, 589.7 9 P. Ehrenberg and K. Schultze, -7. Landw., 1916, a. 37.so W. E. Tottingham, J . Biol. Chem., 1916,24, 221 ; d., i, 460244 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.ever, found that the activity of the nitrogen-fixing organisms wasgreatest in the manure with straw.I n another series of experiments, made to ascertain the effectof different kinds of litter, it was found that all the manures, withand without litter, gained nitrogen in the first four weeks, theincrease with oat straw amounting t o 20 per cent.*1 From thefourth to the eighth week there was a fairly rapid loss of nitrogen.So that in the e'nd all the manures contained less nitrogen thanthey did at the commencement,, the loss being least in the straw-litterea manure.The amount of nitrogen as ammonia, which varied from 0.7 to1.2 per cent. of the total, showed a tendency to increase in thesecond week, after which there was a fall; the proportion of am-monia was, however, never a t all large. The nitrogen soluble inwater, representing 41-48 per cent. of the total, decreased rapidlyduring the first four weeks; the total losses amounted t o 70-90per cent. of the initial amounts, and were somewhat greater in themanure mixed with pine shavings than in the others.Important deposits of potassium salts have been discovered inCatalonia.82 I n one district the amounts of carnallite and sylviniteare estimated a t 2,500,000 and 1,125,000 tons respectively.I n Central Utah, extensive deposits of high-grade alunite havebeen found.83 The mineral contains about 9-5-10 per cent. ofpotash, which can be separated only with some difficulty. Underpresent conditions its exploitation should, however, prove t o bepracticable.N. H. J. MILLER.Within a few hours of posting the manuscript of this ReportDr. Miller was dead. He had worked a t the laboratory as usual,and no one noticed any difference in him; after posting this Reporthe spent the rest of the evening reading and talking with hisfamily. Apparently all was well with him, but two hours aftergoing t o bed he had a sudden pain, and before the doctor couldcome he had passed away. The swiftness of the blow came as agreat shock to his family and his friends. Dr. Miller never soughtpopularity, but he never lostl the friendship of those whom he hadknown in early days: and he was much liked by his youngercolleagues f o r the quiet way in which he was always willing togive up time and take trouble t o help in any way he could.W. E. Tottinghnm, J . I n d . Eng. Chew., 1916, 8, (reprint).E. J. R.82 C. Rubio and A. Rlarin, Rxper. Stat. Rec., 1916, 35, 34 ; 2301. Inst. Gool.83 W. H. Waggaman, U.S. Dept. Agric. Bulletin, 415, 1916.Espaiia, 1914, [ii], 14, 171

ISSN:0365-6217    

DOI:10.1039/AR9161300219

出版商:RSC    

年代:1916

数据来源: RSC

摘要:

RA1I)IOACTIVITY.The H e t e ~ o y e r z e i t y of Cheniiccil E’lerrie7zts.IN the two years 1915 and 1916 since the last Annual Report onRadioactivity much more has been clone t o confirm the generalcorrectness of the new views, wit’h regard t o the nature of thechemical elements and the significance of the periodic law, describedin the Report for 1913. A general description of the point of viewattained niay not be out of place. F o r these new views are farmore fundamental and are far more subversive of the establishedconceptions of chemistry even than was the discovery of the spon-taneous transmutations suffered by the radio-elements. I n the firstplace, they give US the first real knowledge of what constitutes thedifference between one element and another.In the second place,they show t h a t important differences may exist in certain proper-ties, notably the atomic weight and stability, of elements whichare completely identical and homogeneous, judged by all the usualcriteria depended on by the chemist. The analysis of matter intothe so-called chemical elements is indeed only a superficial analysis,and does not imply more than the superficial identity of the atomicstructure. So long as all the known methods of discrimination andidentification depended on properties conferred by the surface orouter shell of the atomic structure, the analysis appeared ultimate.Even the newer X-ray spectra, although a degree more fundamentalthan tlie older methods, are powerless t o reveal any difference what-ever, but the phenomena of radioactivity, in which the innernucleus of tlie atom alone is primarily concerned, has shown matterto be indefinitely more complex than the chemist had hithertorealised.It may now be taken as proved t h a t so long as the netcharge of the nucleus of the atom is the same, the element willshow the definite chemical and pliysico-chemical character associatedwith one or other of the ninety-two places of the periodic table,quite independently of the nature and constitution of the nucleus.Moreover, both with regard to its ordinary light spectrum and itsX-ray spectrum investigated by Moseley, complete identity will be24246 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.found; but the same net charge of the nucleus may result froiiidifferent absolute numbers of positive and negative charges, inwhich case atoms with identical chemical character will show differ-ent atomic masses, the mass being primarily a function, althougli inall probability not a simple additive function, of the number ofpositive charges present.If they are radioactive, they will showalso differences in the manner and the rate of their disintegrationand in the manner and rates of disintegration of their respectiveproducts. Even the same net nuclear charge and the same absolutenumber of the two opposite kinds of charges in the nucleus may beassociated with different nuclear structures, in which case themanner and rate of disintegration will reveal the difference unlessthe atom is completely stable. By f a r the greater number of therecognised chemical elements are stable, and for them the sclepossible criterion for recognising this new complexity is the atomicweight and the few physical properties, such as rate of diffusion,which depend directly on it.I n the radioactive sequence ofelements t-he heterogeneity is much more easily recognised, and, ofcourse, it was first recognised in this field by the existence of theisotopic elements, that is, of elements occupying the same place i nthe periodic table, completely identical in their ordinary analyticaland physico-chemical behaviour. The expulsion in any order of ana-particle carrying two positive charges, and of two P-particles, eachcarrying one negative charge, from the nucleus, for example, leavesthe r e t nuclear charge, and with i t the whole of the chemical char-acter, identical with what i t originally was, although the atomicmass has been lowered by 4 units, by the expulsion of the a-particle.Again, when the series branches, as, f o r example, when some of theatoms expel first an a- and then a &particle, whilst the others expelfirst a P- and then an a-particle, the net nuclear charges and alsothe gross nuclear charge of the two resulting products are identical,but the inner constitution of their nuclei is different, their internalenergy, for example, is different, and therefore their atoms are notidentical.This is a finer degree of isotopy-the chemical character,spectrum, and atomic mass all alike-but the atoms are not thesame, and, if the process of change proceeds further, the difference isrevealed in the manner and rate of the subsequent transfcrmation.If the atoms, on the other hand, are stable, there is no experimentalmethod of disclosing their differences.That is to say, with regardto the ordinary elements, heterogeneity in what has hitherto beenregarded as homogeneous is to be sought for by the atomic mass,but homogeneity in this respect does not preclude the existence ofdifferences beyond the present methods of experiment to establish,but which would be revealed as soon as artificial transmutatioRADIOACTIVITY. 247became possible.with.The atomic weight evidence may first be dealtAtomic Weight, Density, and Spectrum of Lead.The first section of the 1914 Report dealt with the work on theatomic weight of lead from va_rious sources, radioactive and other-wise, and it was then shown that the theoretical prediction that theatomic weight, of lead derived from thorium would prove to beabout one unit higher, and that for uranium about a unit lowerthan the international value, was being borne out.The first workt o be published in this field had reference to the lead derived fromCeylon thorite-not thorianite, a totally different mixed uraniumand thorium mineral often confounde)d with it, even by some of theworkers in this field. I n subsequent work, 30 kilos. of the materialwere hand-sorted into three grades, and from the 20 kilos. in thefirst grade some 80 grams of metallic lead were separated.Thiswas cast in a vacuum into a cylinder and the density determined,ltogether with that of a similar weight of ordinary assay lead simi-larly purified and prepared. The values of D;O were found t o be11.3465 for the ordinary lead, in good agreement with the valuefound by Kahlbaum, Roth, and Siedler-11.3415 for lead distilledin a vacuum-and 11.376 for the thorite lead, which is 0.246 percent. greater. Both samples of lelad were then fractionally distilledin a vacuum, and the atomic weight of the t8wo middle fractionswas determined. The lead was dissolved in dilute nitric acid in aquartz receptacle#, the nitrate evaporated to dryness, and then con-verted into chloride1 by means of a current of hydrogen chloride,the temperature being gradually raised to near the fusion pointof lead chloride, and weighings taken until the weight was constant.The value so found from the ratio P b : PbCl, was, for ordinary lead,207-199, in good agreement with the most recent determinations-207.20 and 207.18.3 The value found for thorite had was 207.694,which is 0.238 per cent.greater. Owing to the researches beinginterrupted by the war, only single atomic weight determinationscould be done.4It will be seen that, as is to be expected on the general view thatthe mass and constitution of the nucleus has no effect on the outerF. Soddy. Nature, 1315, 94, 615.G. P. Raxter and F. I,. Grover, J. Amcr. Ghem. Soc., 1915, 37, 1037;-4.? 1915, ii, 456.0. Honigschmid and (Mlle.) S.Horovitz, Monatuh., 1915, 36, 355 ; A . ,191.5, ii, 635.These results were communicated in two lectures t o the Royal Institution,May 15 and 18, 1915 ; see Engineering, May 28, 1915, and t o Section A ofthe British Association, Birmingham, 1915 ; see Engineering, Oct. 1, 1915,but owing to their incomplete character have not yet been further published248 AXNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.electronic system of the atom, tlie atomic volumes of the two speci-mens of lead are practically the same, the specimen from thoritebeing 0.246 per cent. greater in atomic weight than ordinary leadand 0.238 per cent. more dense. The atomic volume is in each case18.26 a t 20°.An important result attained in this research was t h a t no bismuthcould be found in the whole1 20 kilos.of selected mineral, whicheffectually disposes of the many speculations t h a t bismuth is oneof the ultimate products of radioactive change Thallium, however,was present in easily detectable amount. The supposed a-radiationof bismuth derived from Joachimsthal pitchblende has been defi-nitely shown to be due to admixture1 with ionium.6More extended data are now available for the case of uranium-lead. The preliminary researches of 0. Honigschmid and Mlle.Horovitz,7 which gave the figure 206.73 for the lead from Joachims-thal pitchbtende residues, were undoubtedly influenced by the leadpresent in the crude sulpliuric acid used t o treat the mineral. Thefilial value obtained f o r the lead from 20 kilos.of selected purestJoachimsthal pitchblende, extracted by the use of pure reagents,was 206.405. The lead from the crystallised uranium ore, examinedby Marckwald, from an old primary formation in Morogoro,German East Africa, gave the value 206.046. Almost the samevalue, 206.063, was obtained for the lead from broggerite from&loos, Norway, which contains 79 per cent. of U,O,, 4.5 per cent.of Tho,, and 9.5 per cent. of PbO. Ordinary lead in controlexperiments gave the value 207.180. The arc and spark spectra ofall these specimens of lead showed absolute identity.*I n another and more detailed examination with a grating, of thearc spectra of ordinary lead and of lead from Joachimsthal pitch-blende, complete identity was also found.9 I n the case of two ofthe lines, x 3500 and x 4100, the difference of wave-length must havebeen less than 0.03 if.U., whilst a more accurate comparison ofh 4058 by a Fabry and Perot &don showed t h a t the difference inwave-length could not exceed 0.003 A.U.This is against the theoryof Professor Hicks t h a t the magnitude of the atomic weight entersexactly into the series relationships of spectra.10Further important results on the atomic weight of lead fromAnn. Report, 1914, 286.6 L. Meitner, Physiknl. Zeitsch., 1915, 16, 4 ; A . , 1915: ii, 126.Ann. Report, 1814, 269.0. Hiinigschmid and (Mlle.) S. Horovitz, Monafsh., 1915, 36, 353 ; -4.,T. R. Merton, Proc. Ro?y. SOC., 1915, [A], 91, 19s; A . . 1915, ii, 119.1915, ii, 635.lo ,412n.Report, 1913, 265RADIOAGTIV ITY. 249various uranium minerals have been obtained a t Harvard, and exam-ination has also been made of the density.llLead from the Olary ores, S. Australia, described as Australiancarnotite, and known t o be derived partly from admixed galena,was found t o have a density, a t ZOO, of 11.288 and atomic weight206.34. That from a carefully selected specimen of Norwegiancleveite had a density, a t ZOO, of 11.273 and atomic weight 206.085.For ordinary lead the values 11.337 and 207.18 were found. I n allcases the density is proportional t o the atomic weight, and the valueof the atomic volume is constant a t 18.28, which is practically thesame as t h a t found for thorite lead, 18.26, although the two are notstrictly comparable. Other atomic weights determined were thoseof the lead from &4merican carnotite, 207*004, and from Norwegianbroggerite, 206*122.Thus a t the present time varieties of lead are known varyingfrom 207.7 to 206.05 in atomic weight and from 11.376 to 11.273in density, and in all cases so far examined the atomic volume8 isconstant.The spectra of the different varieties appear to be abso-lutely identical, the one difference in the intensity of the unimpor-t a n t line, 4760.1, noticed in the first sample of thorite leadexamined not having been recorded by other observers.Period, A t o w i c TT'eight, aiid Spectrum of loiiium.Possibly the finest achievement of the period under review hasbeen the comparison in Vienna of the atomic weight and spectrumof thorium with that of the thorium separated by von Welsbachfroin 30 tons of Joachimsthal pitchblende and known t o contain aconsiderable, if indefinite, proportion of the isotope, ionium.Theseexperiments bear out an estimate of the period of ionium recentlyobtained from the1 experiments on tlie growth of radium from puri-fied uranium preparations, which have been in progress for thelast thirteen years, and these results may first be given.12 Thesepreparations showed for tlie first time) beyond doubt t h a t radiumwas being produced from carefully purified uranium, and t h a t therate of production was proceeding as nearly as could be seen pro-portionally to the square of the lapse of time from purification, astheory demands if ionium is the only long-lived element betweenuranium and radium.Thus for a preparation containing 3 kilos.of uranium (element) the growth after three years from purifica-l1 T. W. Richards and C. Wadsmorth, J . Anzer. Chern. A ~ O C . , 1916, 38, 221,l 2 Compare Ann. Report, 1912, 320 ; F. Soddy and(Miss) A. F. R. Hitchins,1658, 2613 ; A., ii, 251, 566.Phil. Mag., 1915, [vi], 30, 209 ; A., 1915, ii, 7?6250 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.tion was 2 x 10-l1 gram of radium, and after six years from puri-fication, 8 x 10-l1 gram of radium. This gives for the period ofaverage life of ionium almost exactly 100,000 years. Any ioniuminitially present would make the estimate1 of the period too low,and it was concluded t h a t 100,000 years might be accepted as aminimum estimate, not far from the real period.Thus there mustbe a t least forty times as much ionium as radium by weight inuranium minerals, t h a t is, a t least 12.3 grams of ionium per 1000kilos. of uranium. The amount of thorium and its isotope, ionium,separated by Welsbach from about 30 tons of Joachimsthal pitch-blende was so small and so intensely radioactive t h a t the questionas to its atomic weight became of the utmost interest. F o r theatomic weight of ionium, calculated from the recent value found foruranium, 238'18,13 by subtracting eight for the two a-particlesexpelled, must be 230.18, and calculated from that of radium, 226.0,by adding four for the a-particle still t o be expelled is 230.0.Themean of these is 230.09, whereas the international value for theatomic weight of thorium is 232.4. The first step was a carefulrevision by new methods of the atomic weight of thorium by theanalysis of thorium bromide,l* analogous to tlie method t h a t hadbeen used f o r uranium. The operations were more difficult, butthe method gave very satisfactory results, and, as in the case ofuranium, a lower value than the1 international figure was obtained,namely, 232.12. The ionium--thorium preparation referred to,treated in an identical manner, gave' the value 231.51, which isnearly 0.6 unit lower. A careful comparison of the spectra of thetwo preparations used in the determinations showed complete iden-tity and complete absence of impurities, thus confirming the earlierconclusion as to the identity of the spectra of these two isotopes.15Ionium and thorium thus furnish the second example' of twoelements differing in atomic weight, b u t spectroscopically and chemi-cally identical.Assuming a mean 230.06 as the true value of theatomic weight of ionium, it has been calculated thatl in the ionium-thorium preparation with atomic weight 231.51 there must bepresent 29.5 per cent. by weight of ionium and 70.5 per cent. of13 Ann. Report, 1014, 272. Further determinat,ions have since been madeof the atomic weight of the uranium separated from Morogoro pitchblende.From the Pb : U ratio, this is some 3.2 times more ancient than the Joachims-thal pitchblende and is practically a pure compound, whereas that fromJoachimsthal contains most, of the known elements.The result found, 238.16,agrees within the limit of error with tho value 238.18 previously found for theuranium from Joachimsthal. (0. Honigschmicl and [Mlle.] S. Horovitz,d4onntsh., 1916, 37, 185 ; A . , ii, 484).l4 0. Honigschmid and (Mllc.) 8. Horovitz, Monatsh., 1916, 37, 305, 335 ;A., ii, 510. Ann. Report, 1912, 321RADIOACTIVITP. 25 1thorium. On this assumption, the life-period of ionium was foundby comparing the a-radiation from a drop of the solution cvapor-ated over a large area with that from a similar known quantityof radium. The life-period so found was 58.1 times that of radium.Tliis is, 145,000 years for the period of average life, or 100,000years for the period of half change,lG and confirms in a satisfactorymanner the period estimated directly from the rate of growth ofradium from pure uranium compounds.I n consequence, it may be calculated that, per 1 gram of radium(element) in Joachimsthal pitchblende, there are 58 grams ofionium and 139 grams of thorium, a total weight of thorium iso-topes of 157 grams.But for the excessively minute proportion ofthorium in this particular mineral, the difference in atomic weightand the identity of spectra of thorium and ionium could not havebeen est.ablished. It follows also that the radium which consti-tutes the1 international radium standard a t Paris and the Viennasub-standards contains so minute a proportion of the isotope, meso-thorium, that the a- and y-radiations of these standards can onlybe affected to the extent of a few thousandths per cent., a quantityfar below the limit of accuracy of radioactive measurements.The Definition of the Atom and the Element.The foregoing account contains most of the important new factshaving reference to the existence of isotopic elements, but a largenumber of r&um& and theoretical papers have appeared, dealingwith the new conceptions and the progress made in our knowledgeof the nature of matter generally. First, however, may be men-tioned a thoughtful pap’er on the changes in chemical nomenclaturenecessitated by the widening of our outlook.17 The subject is treatedhistorically, and the gradual evolution of the meaning attached tothe terms element and atom traced from the time of Boyle.Dalton’satomic theory and Boyle’s conception of elements together led tothe point of view that there are as many kinds of absolutely similaratoms as elements, a view that is now no longer true. Panethrecommends retaining the idea of elements previously held andspeaking of all the isotopes of one element, o r of a mixture ofthem, still as one element. The light and X-ray spectra wouldthus still remain the distinguishing criteria of a chemical element,which would be defined as a substance that cannot be decomposedinto anything simpler by any chemical process. Two elementswould be denoted by the same name if, once mixed, they cannotbe separated by any chemical process. Atoms, on the other hand,R.Mcyer, Monatsh., 1916, 37, 347, A., ii, 511.F. Paneth, Zezksch. physikal. Chem., 1916, 91, 171 ; A . , ii, 2402.52 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.would be defined as the particles of matter, unchanged by chemicalanalysis, which represent the limit of chemical decomposition.Elements made up of the same kind of atoms would be defined as((pure,” and those comprising different kinds of isotopic atoms as((mixed.” It must be confessed, when one tries to carry out theselogical rules into practice, difficulties must arise. Thus radium andmesothorium are the same element, and must have the same name,and similarly with ioniuin and thorium. The facts will make, it isto be feareld, as much havoc among these new rules as it hasalready done among the old.The whole trouble arises from thefact t h a t the words atom and element imply the idea of ultimateconstituents, and we know now that our analysis of matter cannever be ultimate.Theories of A toniic Structure.Physicctl Properties of Isotopes.-After the discovery of thedifference of density between tlie different varieties of lead, thegeneral question as to the physical and chemical properties ofisotopes received theloretical attention. Lindemannl8 concludedthat if the atomic volumes of isotopes were the same, their elasticconstants, vapour pressure, and melting points should differ, and hepredicted that the le’ad from thorite should melt 1 * 5 O above themelting point of ordinary lead, a prediction which, like manyothers, it lias not yet been found possible to test.He regards theforces of attraction and relpulsion between the atoms, the inter-action of which results in tlie solid state, as originating in tlienucleus. Apparently these conclusions are opposed to those drawnby Bohr.lg Another examination from the point of view ofBolir’s theory has been made by Fajans,20 who concludes that a tabsolute zero tlie atomic volumes of isotopes should be identical, butshould difTer progressively as the temperature rises, the differenceamounting to less than 1 / 10,000tli part a t the ordinary temperaturefor the isotopes of lead. The specific heats should, however, bedifferent, the greatest difference amounting in the case of the leadisotopes to 0.75 per cent.It is clear t h a t the further examinationof the physical properties of the isotopes of lead will prove a verysearching means of testing some of the) newer pliysico-chemicaltheories of the solid state.A tom BiriZdiug.-A series of papers has been published giving aP. A. Lindemann, Naewe, 1916, 95, 7.l9 British Association Meeting, 1915. Discussion in Section A. See2o I<. Fajans, Arbeiten cius r l e ? ~ Geheit rlcr Physik, Afrith. Cl~emie, 623 ;A’ngineeriny, Oct. 1 , 1915.-4., ii, 400RADIOACTIVITY. 253comprehensive survey of much of the modern work and its detailedbearing on chemical problems.21 The possible existence of isotopesmay account for some of the largar departures of the atomic weightsfrom whole numbers, such as those of magnesium and chlorine.Onthe modern theory, however, t h a t mass is wholly due to electro-magnetic inertia, it is not t o be expected t h a t the mass of an atomshould be the exact sum of the separate masses of its constituentsub-atoms, although the deviations on this account are probablyquite small. The problem of mutual electromagnetic mass, as it iscalled, has also been considered by other workers.22 Indeed, in1910, Silberstein solveid the problem rigorously for the two physi-cally important cases, either when the electric charges are entirelyseparated or when one is entirely within the other. His expres-sions, which are contained in the reference cited, make the differ-ence between the masses of the separate components and that ofthe complex a relatively simply function of the radii of the electriccharges and of their distance apart.Most important informationas to these atomic dimensions could be immediately obtained if theatomic weights of the various products of a disintegration seriescould be determined to the very high degree of accuracy required.Nicholson, with an approximate solution, has already concluded,from such data as were available for thorite lead, that the meandistance apart of the a-particles in a thorium atom is of the sameorder as the radius of the electron.The mean departure from whole numbers, in terms of hydrogenas unity, of the atomic weights of the first twenty-seven elements,excluding hydrogen, is nearly a constant percentage, 0.77. Foroxygen, it is exactly this, and therefore the approximation t o wholenumbers of the atomic weights of these elements, on the basis ofoxygen as 16, is very close.The chance of this being accidental iscalculated a t one in fifteen million. This 0.77 per cent. is regardedas the packing effect, or loss of mass when the hydrogen nuclei arepacked together to form a heavy atom. As an illustration merely,of no practical application, it is calculated t h a t a positive andnegative electron would lose 0.77 per cent. of their separate massesif caused t o approach to a distance apart four hundred times theradius of the positive electron. From nickel onward no tendencyto approximate to whole numbers seems t o exist, for the mean21 W. D. Harkins and E.D. Wilson, J . Arner. Chem. SOC., 1915, 37, 1367,1383, 1396* ; 1916, 38, 169 ; A., 1915, ii, 543, 544, 544* ; 1916, ii, 241 ;Phil. Mag., 1915, [vi], 80, 723, A., 1915, ii, 814. The paper asterisked is asummary of modern work, especially valuable in its interpretation of Nichol-son’s theories.22 L. Silberstein, Phil. &lay., 1015, [vi], 30, 370; J. W. Kicholson, ibid.,6 5 9 ; Proc. Physical SOC., London, 1915, 27, 2 1 7 ; A . , 1915, ii, 404254 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.departure from integers for the thirteen best determined atomicweights between nickel and barium is 0.25 unit, as it should be 011a purely chance distribution.I n common with van der Broek,23 the helium nucleus is picturedas built up out of four hydrogen nuclei and two negative nuclearelectrons.The authors regard nearly all the (‘ packing effect” t ooccur in the formation of these helium nuclei, the aggregation ofthese into more complex atoms not producing any noticeable changeof mass. Along these lines, the constituent helium nuclei beingregarded as much more stable than the complexes formed out ofthem, is to be sought the explanation of why helium and nothydrogen is expelled in the disintegration of radio-atoms. Vander Broek, suggests the further aggregation, within the complexatoms, of four helium nuclei and two nuclear negative electrons,this complex being the oxygen nucleus. He ascribes radio-activity to the disintegration of these with successive expulsion offour a- and two &particles, a common sequence in the radioactiveseries.This represents a change of 16 units of atomic mass for6 of atomic number, and this, he points out, is the average for allthe elements between magnesium and thorium, which differ by 208units of mass and 78 units of atomic number, that is, as 16 isto 6.Periodic Law.-The fifth paper of the series deals with theperiodic law and its representation by helical models. Many usefuldata, relative especially to the physical properties of the elements,are here collected together. The authors are alive to the essentialchanges of outlook effected by the recent advances on the repre-sentation of the Periodic Law, but their model offers little ofnovelty or advantage over earlier efforts, although discussed inmuch greater detail.I n one important point they seem to be atfault. They divide the Periodic Table into three cycles of 42, 6?,and 82 elements, and each cycle into two periods each of 8, 18,32 elements respectively, the last part of the last very long periodof 32 elements being missing, and they argue that this representsa numerical expression of some function appertaining t o atomicstructure; but the series from the emanation to uranium runs likethat from krypton to molybdenum, and not like that from xenon totungsten, wherein the rare earths are interpolated.24 Although thelast very long period is incomplete, i t runs far enough t o makethis division of the table into cycles of 42, 62, 8 2 elements fanciful.Again, the eighth group is still only accommodated in their model23 A.van der Broek, Physika7. Zcitsch., 1916, 17, 260 ; Nature, 1916, 97,24 Soddy, “ Chemistry of the Radio-Elements,” Part 11, p. 11, 1014.479 ; A . , ii, 465RADIOACTIVITY. 255on the old supposition that “ a catastrophe of some sort seems totake place here,” whereas these elements with their gradual changeof properties are as significant and marked a feature of the PeriodicLaw as is the intensely abrupt change of properties that occurs oneither side of the zero group.Stellar Elements.-Fraunhofer’s discovery of the dark lines ofthe solar spectrum, the discovery of helium, and the theory of theevolution of the heavy elements from the lighter ones that hasbeen advanced by Lockyer as a consequence of the study of thelife-history of the stars, have made chemists familiar with thefruitful bearing of astronomical observations on their own science.From such sources, aided by mathematical reasoning applied to anew theory of atomic structure, Nicholson has formed some tenta-tive conclusions of great interest to chemists 25He regards the terrestrial elements as differing in character andatomic structure from the elements coronium2 nebulium, andothers, the existeiice of which, like that of helium, has beeninferred from unidentified lines in the solar corona and in nebuh.These lines appear to him to originate from earlier and simplerevolutionary forms of matter from which the terrestrial elementshave developed.It is significant that there seems to be no roomin the Periodic Table, as we now understand it, for any of thesestellar elements. Nicholson’s model atom is of the Rutherfordtype, in which the electrons revolve in orbits round a centralpositive nucleus, but in his calculations, unlike Bohr, he does notdepart from the classical mechanics or introduce the quantumhypothesis.The energy of the spectrum is not derived frominternal atomic energy, but from external sources, and on thispoint there can scarcely be any reasonable doubt that he is correct.I n spite of its great initial successes in calculating correctly themagnitude of the Rydberg constant, and in correctly ascribing thePickering series of lines t o helium rather than to hydrogen, Bohr’stheory does not seem to have been generally so successful. I n theNicholson atom, the vibrations from which the spectral linesoriginate are perpendicular to the plane of the ring of revolvingelectrons. The strongest and first class of vibrations are due t othe entire ring vibrating as a whole, keeping parallel to its posi-tion when not in vibration.In the second class of vibrations, thering vibrates in halves, that is, there are two nodes and two creststravelling round the ring. In general, there are as many classesof vibrations possible as there are electrons in the ring, although25 Nicholson’s earlier theories were fully discussed in the Annual ReportThe present account is derived mainly from the third paper for 1911, 269.in the series referred to, by Harkins and MTilson256 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.the higher classes are probably too feeble to detect. For thenebulium atom Nicholson postulates a nucleus of four positivecharges. The chief line attributed to this element is h=5006*9,and if this constitutes what has been termed the first class, thecalculated wave-length of the second class should be 4367.0, whichcorresponds with the strong observed line h=4363.4.The nextshould be 4352.9, and it was only after its wave-length had beencalculated that a line 4352.3 was photographed at the LickObservatory. A search of older photographs showed its existenceon plates taken several years before, but it had escaped observa-tion owing to its feebleness. Again, two lines formerly attributedt o nebulium could not be obtained from the structure of the atompostulated, and these, simultaneously with this result, were shownby Wolf, of Heidelberg, to originate in a different part of thenebula from the rest.Similarly for model atoms of 2, 3, 5, and 6 charges, by assumingone line to be due to the first.class of vibration, the wave-lengthsof the other classes of vibrations have been calculated and corre-spondence found with observed lines. Thus twenty-one out of thetwenty-seven lines of the solar corona have been accounted for bythe vibrations of a model atom of five charges. The atomic weightsof these stellar elements can only be very indirectly and unsatis-factorily inferred, but the values given are 0;082 multiplied by thesquare of the nuclear charge.Thus protohydrogen is 0.082,nebulium 1.3, protofluorine or coronium 2.1.This work thus suggests the existence of types of matteraltogether unknown upon the earth, and for which no place existsin the Periodic Table. The difference between the two kinds isprobably t o be sought in the constitution of the nucleus of theatom. We know from the expulsion of &particles and the exist-ence of isotopes that negative electrons do form an integral partof the nucleus of the radioactive atoms, and that the nuclearcharge or atomic number is the difference between the number ofpositive and negative electrons. Nicholson’s idea seems to be thatthis is true of all terrestrial elements, including hydrogen, andthat it constitutes the difference between these elements and thestellar elements. On this view, then, the hydrogen nucleus is nota single positive electron, but a combination of several with anumber of negative electrons one less.This in turn involves asimpler ultimate unit of atomic structure than hydrogen, but isnot necessarily inconsistent with the conception of nuclei, ofhelium, and of hydrogen as penultimate nuclear constituents of themore complex atoms known on the earth.Cause of Atomic Disintegration.--For the first time, an attempRADIOACTIVITY. 257has been niade to frame a theory of the process resulting in atomicdisintegration.26 Hitherto the barrier has been the exponentiallaw of change, which is the law of pure chance, that out of adefinite number, Q, of atoms a definite fraction, AQ, break up inthe unit of time, independent of every consideration whatever.On this theory, the cause of instability is that N separate particlesin the atomic nucleus should all pass some critical position in theshort time 7, 7 being the time taken f o r a strain (regarded as asound-wave) to traverse the nucleus.An analogy is given whichhelps to make the theory intelligible. A number of smallimpulses applied to a pendulum will result in a maximum displace-ment if all are applied while the pendulum is moving in one direc-tion. The constant A, then, appears as the probability ofLT particles being all in the critical region in the short interval 7.This is (rv)N, where v is the frequency or number of times theparticle passes the critical region per second.It is shown thatN is 1.5 B, where B is the constant of the Geiger-Nuttall relationbetween A and the range R of the a-particle expelled, namely,logh =n + 13 log R . N is equal to 80, which, as the atomic numberis between 80 and 90, indicates that nearly all the free positiveparticles in the nucleus must conspire t o effect its disintegration.The radius of the nucleus, evaluated on this theory from theconstant A , agrees with that found by Rutherford for the goldnucleus.High- f re p z L e n c y S y e c t rct of t h e Ele rti r: 11 t s.As previously pointed 0 ~ t , 2 7 the discovery of the regular reflec-tion of X-rays from crystal surfaces has been the experimentalmeans of two distinct advances, the elucidation of crystallinestructure and the determination of the wave-length of the X-rays,the latter only falling within the province of this Report.It wasby this means, some six months ak'ter the elucidation of the natureof a- and &changes had proved that the consecutive places a t theend of the Periodic Table correspond with unit difference of atomiccharge, that Moseley was able to extend the conception towardsthe beginning of the table as far as aluminium, and to call theroll of the elements, which confirmed in so striking a manner theaccumulated labours of chemists since the time of Robert Boyle.Numerous extensions and confirmations of this fundamental workfall to be recorded.I n a long series of papers,2* the L-series of spectra, for which26 F.A. Lindemann, Phil. Mug., 1915, [vil, 30, 5 6 0 ; A.. 1915, ii, 720.27 Ann. Report, 1913, 273.REP.-VOL. XIII. KCompare a!so A4. Debierne, Ann. Phpique, 1916, [is], 4, 323 ; A., ii, 168.&I. Siegbahn and E Friman, PAysikuZ. Zeitsch., 1916, 17, 17 ; PhiE. May.258 ANNUAL REPORTS ON THE PROGRESS O F CHEMISTRY.Moseley observed five line groups and measured three,Zg has beenmuch extended. Fourteen groups of lines are now recognised inthe L-series, and the elements examined have been extended asfar as zinc in one direction to uranium in the other.30 Indeed,very few of the elements now have not had their atomic numberdirectly determined by measurement of the frequency of one ormore of the lines in one or more of the various series. Animproved vacuum spectrograph has enabled the L-series t o befollowed down t o wave-length 12-35 A ., whereas the greatest wave-length hitherto measured was 8.4 ;\.Only for three lines (al,a2, and p 2 ) is the linear relation between atomic number andsquare root of the oscillation frequency exact. For the othersthe curves are slightly convex to the atomic number axis. Theatomic number of uranium, the last element, is, as previously pre-sumed, 92, thorium being 90, bismuth 83, lead 82, thallium 81,mercury 80, and gold 79, while it is claimed that the values 84and 85 for polonium and radium have been experimentally estab-lished. With regard to lead, an interesting table is given31 com-paring twelve lines of its L-series spectrum with fifteen lines ofthe soft y-ray spectrum of its isotope, radium-B, as given byRutherford and Aiidrade.3' Ten of the lines are nearly identicalin wave-length, thus confirming still further the identity of theX-ray spsctra of isotopes. Tellurium and iodine have the atomicnumbers 52 and 53 respectively, as the periodic law requires.I nthe K-series, four lines have been measured for the elementschromium to germanium, and, in the most recent work, investiga-tion has been pushed as far as sodium, with results in full accordwith Moseley's work. Lastly, a new series, the M-series, comprisingsix or seven rays, softer than the L-series, but very similar incharacter to this and to the I<-series, has been found, and thewave-length measured for the elements uranium to gold.Thisseries also obeys the linear relation over the range studied. Thusthe X-ray spectra are losing their pristine simplicity, but thetheoretical interpretation of the various series should still be child'splay t o the mathematician in comparison with that of light spectra.The atomic numbers of the rare earths do not differ from thosepreviousIy given.33 Numbers 61 and 72 are still vacant, but the1916, [v;], 31, 403 ; &4., ii, 167 ; Physikal. Zeitsch., 1916, 17, 48, 61 ; A . ,ii, 205, 277 ; M. Siegbahn and W. Stenstrom, ibid., 1916,17, 318 ; A . , ii, 509 ;M.Siegbahn, Ber. Deut. physikal. Ges., 1916, 18, 39, 150, 278 ; A., ii, 462,462, 463 ; also Ann. Physik., 1916, [iv], 49, 611, 616 ; A., ii, 362 ; Phil.Mag.,1916, [vi], 32, 39 ; A., ii, 405.30 E. Friman, Phil. Mug., 1916, [vi], 32, 4 9 7 ; A . , ii, 689.31 Phil. Mag., 1916, [vi], 32, 49.32 Ann. Report, 1914, 275. 33 Ibid., 28028 Aim. Report, 1914, 278RADIOACTIVITY. 259others have all been measured except No. 69, ascribed to thulium.Urbain3.1 states t h a t celtium has not yet been discovered by thismethod, t h a t there is only one thulium, one erbium, and twoytterbiums-neoytterbium and luteciurn. It is to be hoped thatagreement will now be reached as to the names of these elements,as several of them, having been discovered independently, havealiases. A number of other papers on this subject have alsoappeared.35The rays from the Coolidge X-ray tube with a tungsten anti-cathode have been subjected to an interesting examination t odetermine the maximum frequency of the X-rays emitted underdifferent constant voItages.35 The hope previously expressed,37 thatthis tube would enable X-rays as penetrating as the ?-rays ofradium to be artificially generated, has now been definitely dis-pelled.The penetrating power and frequency of the X-raysreaches a maximum a t 143,000 volts, and does not sensibly increasebeyond this when the voltage is raised to 1'70,000 volts. Theminimum value of the absorption coefficient, p(cm.) -I, Al, is 0.39,as compared with 0.115 for the penetrating y-rays of radium-C.For lead the value is 23, as compared with 0.5 for radium-C, and3 mm. of lead afford practically complete protection against themost penetrating X-rays that the Coolidge tube can produce.Thelimiting frequency is probably determined by the highest frequencythat exists in the atom, i n this case, of tungsten. It is anticipatedt h a t by the use of a uranium anti-cathode, a limiting penetratingpower of p =0'23, instead of 0.39, in aluminium could be attained.Plaiick's quantum theory appears to hold for the relations betweenfrequency, v, and voltage, E, for small voltages, but for highervoltages, instead of Izv=E, the relation assumes the formhv=E-cE2, where c is a constant. This formula does not holdbeyond the maximum frequency given by E=1/(2c). The voltagerequired t o excite the most penetrating radiation is about twicewhat is t o be expected on the quantum theory, and the eficiencya t high voltage is only about 1/500th. If one-half of the energyof each contributing electron appears as radiation, this indicatest h a t only one in some two hundred electrons contribute t o theradiation.I n this work there was no evidence that the radiationfrom the tube could be analysed into definite characteristic radia-34 Obituary notice of H. G. J. Moseley, Proc. Roy. Soc., 1916, [A], 93,xsvii.35 1. Malmar, PhiZ. M n g . , 1914, [vi], 28, 787 ; A . , 1915, ii, 2 ; 32. dr, Broglie,Compt. rend., 1916, 163, 57 ; A., ii, 509 ; J. Barnes, Phil. Alag., 1916, [ \ T i ] ,30, 3 6 8 ; A., 1915, ii, 658.3G Sir E. Rutherford, J. Barnes, and H. Richardson, ibid., 339, 361.37 A m . Report, 1314, 277.K 260 ANNUAL REPORTS ON THE PROGRESS OF CIIEMISTRY.tions, although the initial potential, 10,300 volts, a t which X-raysbegan to be generated corresponds well with t h a t required t oexcite the L-characteristic of tungsten, and the most penetratingrays correspond with the shortest wave-length component of itsI<-characteristic radiation.On the other hand, de Broglie hasfound it possible with a Coolidge tube to follow the absorptionband of an element a t least as far as bismuth, and considers thathis results put beyond doubt the presence of radiation far morepenetrating than the X-radiation of its tungsten anti-cathode.a-, 0-, and y-Radiutio,is.a-Rays.-Hitherto, the fastest known a-rays have been those ofthorium-C', with a range of 8.16 cm. in air a t N.T.P. A smallnumber of bright scintillations, undoubtedly due to a-rays, froma strong preparation of the active deposit of thorium, were observedon a zinc sulphide screen after the passage of the a-rays throughthe equivalent of 10.7 cm.of air a t N.T.P.39 Further investigationsdisclosed t h a t two new sets of a-rays were probably present, one-third of range 9.7 cm. and two-thirds of range 10.7 cm. in air a tX.Y.P. The division is just the same as for the ordinary a-raysof thorium-C, which gives two sets of different range, one-thirdwith range 4.55 and two-thirds with range 8.16 cin. These newa-rays certainly come from the active deposit of thorium, for theydecay a t the same rate as the other a-rays, but it is not yetdefinitely proved t h a t they come from thorium-C.It is suggestedthat the a-particles of 9.7 cm. range accompany those of 4.55 cm.range, and those of 10.7 cm. range accompany those of 8.16 cm.range. For each a-particle in the two new sets, 10,000 in the twoold sets are emitted. The Geiger-Nuttall relation indicates thatthe period of average life of the atoms yielding these two new setsof a-rays must be 10-13 and second respectively, and theirvelocities are estimated to be 2.18 and 2.26 ( x 1 0 9 cm. per sec.).A new determination of the velocities of the other two sets ofa-rays from thorium-C gave 1-714 and 2.060 ( x 1 0 9 cm. per sec.),and this indicates t h a t the range of the slower a-particle shouldbe about 4.70 cm. instead of 4.55 a t iV.T.P.40 A determinationof the velocity of the a-particle from radium-A gave 1.69 ( x 109 cm.per sec.), in good agreement with that calculated from the range.41It has now been definitely shown e that the emission of 8-rays from39 Sir E.Rutherford a,nd A. 13. Wood, Phil. Maq., 1916, [vi], 31, 379 ; A . ,i j , 282.4 0 A. R. Wood, i b i d . , 1915, [vi], 30, 703; A . , 1915, ii, 814.42 J. McLennan and C. C,. Found, ibid., 30, 491 ; A . , 1915, ii, 712. Coin.N. Tunstall and W. Rfakower, ibid., 29, 259 ; A . , 1915, ii, 80.pare Ann. Report, 1912. 301RADlOACTIVITY. 2G 1metals bombarded with a-rays is, like the photoelectric effect,dependent on a surface gas film, and in the complete absence ofthis no emission occurs.Several microphotographic investigations of the tracks ofa-particles in photographic films have been made,43 and the hope isexpressed that with a suitable plate (Wratten and Wainwright'slantern plate being alone found suitable) all the phenomenahitherto studied by the scintillation method can be more con-veniently studied by this means.Certainly such a method wouldhave many advantages. The halos obtained when an active needlepoint is made to touch a photographic plate have under the micro-scope many of the characteristics of the pleochroic halos. Themethod has been used to investigate the straggling of a-particlestowards the end of the range, with results much closer in accordwith a theoretical formula, calculated by Bohr, than earliermeasurements by other methods,Twoattempts have been made, by Wilson's method, to photograph thetracks of the H-particles known to be produced in the passage ofa-rays through hydrogen.The first45 was not attended withsuccess. I n a large number of beautiful photographs of a-raytracks in hydrogen, no evidence whatever of the production ofH-particles was obtained. I n the second,*R the separation of thepath of the a-particle, a t the point of its collision with thehydrogen nucleus, into two distinct paths, corresponding with therecoiling H-particle and the deflected a-particle respectively, wasclearly shown, and the number observed were in satisfactory agree-ment with the Rutherf ord-Darwin formula.On the other hand, Mar~den,~7 who first observed these 11-par-ticlea, in attempting a quantitative verification, by counting thescintillations they produce, in an apparatus similar to t h a t used t ostudy the scattering of a-particles, observed effects several timesgreater than the formula leads to.It was then found t h a t thea-ray tube and emanation alone were emitting N-particles, withoutany layer of paraffin-wax, which was put on as a convenient formof hydrogen atoms. The source of these is doubtful, as i t seemsunlikely that there was sufficient hydrogen in the glass tube, or anyH-P'crrticZcs.44-This subject is in rather a confused state.43 S. Kinoshita and H. Tkeuti. Phil. Mag., 1915, [vi], 29, 420 ; R. R. Sahni,ibid., 836 ; H. Ilieuti, ibid., 32, 129; W. Makower, ibid., 222.41 Ann. Report, 1914, 274.45 J. C. Mc1,ennaii and H.V. Xercer, Phil. -?tug., 1915, [vi], 30, 676.4G D. Rose, Physikal. Zeitsch., 1916, 17, 388 ; A.. ii, 547. Unfortunately,the ahstract only is available to the writer, who has not seen tho actualphotographs.47 E. Marsden and W. C . Lantsberry, Phil. Mag., 1915, [vi], 30, 240262 ANNUAL REPORTS ON THE PROGRESS OF C‘HEMISTHT.condensed water filni present, to produce them, and the suspicionis crested t h a t they are emitted by the radioactive atoms them-selves. This seems to contradict the results of tlie special researchmade for radiant particles, differing either in mass or charge fromthe a-particles, in the rays from the radium emanation,@ but in thelatter it is stated t h a t the nuinber is certainly less than 1 in 10,000a-particles, and possibly the two statements are not inconsistent.P-Rnys.-Two sets of P-rays, of velocity 0.51 and 0.47 times t h a tof light, hitherto attributed to thorium-5, have been shown t o bedue t o radio-tlioriuin, which in this respect is now analogous t oradio-actiniurn.49 The paper contains useful information on thedeposition of filnis of thorium-X and radio-thorium on short, finewires.The two sets of &rays emitted by radium-D, velocities 0.39and 0.37,sO have been the subject of an interesting examination.Hitherto, on account of their feeble penetrating power, which is lessthan t h a t of the a-rays, they have only been detected photographi-cally by means of their magnetic spectrum. After an extremelycareful purification from radium-3, which gives penetrating &rays,and radium-F, which gives a-rays, radium-D was found t o be givingP-rays capable of detection by the electroscope, which were reducedin intensity t o 10 per cent. of the initial value by 0.0035 mm.ofaluminium. The absorption-coefficient is 5500 (cm.) -1 Al, whichagrees almost exactly with the value, 5067, f o r cathode rays of36 per cent. light velocity, as measured by Becker. The experimeiitsfavoured the view t h a t tlie absorption of &rays is primarily dueto tlie completed stoppage of the individual particles in singleencounters with atoms, rather than to the gradual wearing down ofthe velocity of the whole beam, alternatives that have previouslybeen fully discussed.51 With regard to the absorption of homo-geneous 8-rays in aluminium, the absorption curve is nearly linearand the rays have a definite “range,” varying from 0.025 cm.f o rrays of Hp=1930, velocity 0.75, t o 0.5 cm. f o r rays of Hp=11,500,velocity 0.99 times t h a t of light. This is due t o the chance balanc-ing of tlie scattering and the diminution of velocity, for in paper,where the scattering is smaller, the absorption curve is concave, andin platinum, where it is large, convex to the origin.52y-Bnys.-Experiments on the excitation of y-rays by P-raj s inplates of iron, nickel, copper, and zinc showed t h a t the y-raysexcited were similar in abso~rption-coeffic:ent t o Earkla’s charecter-4 5 Ann. Report, 1914, 27k.4 9 0. Baeyer, 0. Hahn, aiid L. Meitner, PhysikctZ. Zeitsch., 1915, 16, G ; &4.,50 Ann.Report, 1911, 2 7 8 ; L. Meitner, ibid., 272; -4., 1915, ii, 663.51 -4nn. Report, 1007, 315. ‘‘ R. W. Varder, Phil. M u g . , 1915, [vi], 29, 725 ; A , , 1915, ii, 4011915, ii, 127RADIOACTIVITY. 263istic secondary S-radiations froin these metals, and, in silver andtin, with Barkla's K-radiations of these meta1~.~3 The absorption ofthe y-rays of radium-B and -C in lead shows no such anomaly, dueto the atomic numbers of these elements and of lead being similar,as might have been expected from Barkla's work on the absorptionof X-rays. Eiglity-five per cent. of the ionisation due to radium4is from the well-known penetrating rays, ~ ~ 0 . 5 (cm.)-I, Pb, and15 per cent. to rays of apparently exactly the same character as they-rays of radium-B.The latter comprises three types, with p inlead, 1.5, 6.0, and 46 respectively contributing 12, 26, and 46 percent. of the total ionisation.5'Rcrdioactive Recoil.Once thel charge carried by the a-particle was a debated problem.55Was it intrinsic, or was it simply the result of its encounter withtlie first atom in its way? It was very certain that an atom travel-ling a t such a speed would be ionised itself, even if initially un-charged. After many researches, it was finally decided that, sofar as experimental tests could show, the charge was intrinsic, andnow we know t h a t it is the loss of tliese two charges from thenucleus in an a-ray cliaiige which is the cause of- the shift of twoplaces in the Periodic Table.Exactly the same question has nowbeen asked of the recoiling particle,5G which ordinarily carries asingle positive cliarge,57 a i d here also tlie tlieoretical importance oftlie question is considerable.5e The recoil of radium-D fromradium-6' was clioseii, and here it was a t once found that in a suffi-ciently good vacuum the recoiling particle is uncharged. It gainsits charge by collisions with the molecules of the gas in its path.I n a vacuum a t a pressure measured by the Knudsen absolutemanometer to be 0.6 dyne per cm.2 (4.5 x lo-' mni. of mercury)-six hundred times, i t may be remarked, t h a t measured by theMcLeod gauge-no charge was carried by the recoiling atoms. Asthe pressure rose, the charge acquired increased and, a t sufficientlyhigh pressure, equalled t h a t carried by the a-particles, showingthat the radium-D atom can acquire multivalent charges.Sincea t atmospheric pressure a univalent charge is carried, it is clearthat, as in the case of the canal-rays, successive recombinaticns and53 (Miss) J. Szmidt, Phil. Mag., 1915, [vil, 30, 220 ; A4,, 1915, ii, 721.54 H. Richardson, PYOC. R o ~ . SOP., 1915, [A], 91, 396; A . , 1915, ii, 401.55 Ann. Report, 1904, 256 ; 1905, 302 ; 1906, 344 ; 1907, 315.66 L. Wertenstein, Comnpt. rend., 1915, 161, 896 ; A., ii, 69.57 Ann. Report, 1910, 272 ; H. P. Walmsley and W. Makowor, Phil. Mag.,5 5 Ann. Report, 1913, 281.1915, [vi], 29, 253264 ANNUAL REPORTS ON THE PROGRESS OF CHEMISTRY.ionisations occur along the path.It would not be safe to generalisefrom the case of radium-]), for it is produced in the excessivelyquick change of radium-C':, which itself, a t the moment of itsdisruption, may still carry a positive charge acquired a t its birth,10-11 second earlier, by the expulsion of a P-ray from radium-C'.This might lead t o the expectation t h a t in other cases the recoil-atom will be found negatively charged. The research opens outa whole series of questions as to the conservation of electric chargesduring radio-active changes, and further results will be awaitedwith interest.Attempts have been made t o detect a photographic action bothof radium-G recoiling from polonium,sg and of radium-]) fromradium-C,Go using Schumann plates in which the amount of gelatinin the film was reduced t o a minimum.No photographic and nophosphorescent effect on zinc sulphide could be detected in the firstcase, but, in the second, a weak photographic effect, attributed tothe recoil atoms, was obtained. A line in the magnetic deflectionexperiment was found nearly midway between that due to thedeflected and undeflected a-particles of radium-C, and it was con-cluded t h a t the atoms of radium-]) are projected with a singlepositive charge, as is the case with radium-]) recoiling fromradium-A. The experiments were done in a vacuum, but it is clear,from Wertenstein's results, already noted, t h a t the charge carriedby the recoiling atom of radium-11 varies with the pressure ofthe gas.A method of determining the ratio of the " ranges " of the recoil-ing atoms of radium-B and -D respectively from the efficiency ofthe recoil of the latter from a plate on which radium-A had beendeposited indicated a greater difference in their ranges than inthose of the corresponding a-particles of radium-;l and radium-6'respectively which accompany these recoiling atoms.61Chemical L4 ctioizs of t h e Radicttioiis.-The spontaneous reactionbetween hydrogen arsenide and oxygen is accelerated by P- andy-rays, and, instead of the arsenic being liberated as such, it isoxidised to arsenious acid, 2AsH3 + 30, = 2€13As03.62 The velocityof the esterification of acetic acid is scarcely increased by the actionof the radium rays, but' ethyl acetate is decomposed by the pene-trating rays, primarily with the formation of an acid and an uii-saturated hydrocarbon.The oxidation of acetone t o acetic acid by5 9 A. B. Wood and ,4. T. Steven, Proc. Physical Soc., London, 3915, 27.6" A. B. Wood and W. Rfakower, Phil May.. 1915, [vi], 30, 811 ; A., ii, 6.62 H. Rerltleben and G. Lockcmatin, Zeitsch. anol-9. Chein., 1915, 92, 145 ;189 ; A . , 1915, ii, 403.W. Maltower, ibid., 1916, [Ti], 32, 226; A . , ii, 547.A . , ii, 209RADIOACTIVITY. 265radium rays is detectable, but, as is usually the case, is very slightcompared with similar action brought about by ultra-violet light.63An investigation of the union of hydrogen and chlorine, under theinfluence of a-particles from radium emanation, showed t h a t thereaction is of the first order, is not affected by the hydrogen chlorideformed, and is retarded by oxygen a t all stages proportionally tothe oxygen concentration.As regards the kinetics of the combina-tion, the reaction showed complete parallelism to the photochemicalreac tion.G4I n a series of experiments on colloidal solutions, the penetratingrays of radium were found to precipitate electropositive, but notelectronegative, colloids. Native, well dialysed albumin is changedand finally coagulated by the rays, which produce also a loweringof the coagulation temperature and increased precipitability byalcohol. Salts exercise a protective action.65i1-e IJI TT'ork o I L t h e Rtrrlio el P ttl P 11 f c.A number of more extended accounts of collected investigations,previously published in brief, have appeared dealing with (1) theradio-active decomposition of water ; (2) the production of heliumfrom radio-active substances ; (3) the atomic weight of the radio-active emanations; and (4) the deposition of the active deposit ofradium.66Thorium.-The life-period of thorium has been determined as1.77 x 1010 years for the period of half-change, 2.56 x 1010 years forthe period of average life, from the ionisation of films of thoriumoxide of vanishing thickness.67Rndi7cnz.-The life of radium has been redetermined by Bolt-wood's method,G* by separating the whole of the ionium in auranium mineral and determining the rate of growth of radiumfrom it in terms of the equilibrium amount present in the mineraLG0Carefully selected specimens of uraninite (N.Carolina), cleveite(Norway), and broggerite (Norway) were used. The ionium wasseparated by addition and separation of thorium, several separationsbeing carried out, and the last kept distinct for a blank test, toG3 A. Kailan, Monatsh,. 1914, 35, 859; A . , 1915, ii, 663.64 H. R. Taylor, J . Amer. Ckem. Soc., 1915, 37. 2 4 ; A . , 1915, ii, 80.G,5 A. Fcrnau and W. Pauli, Biockem. Zeitsch., 1915, 70, 426 ; A . , 1915,i , 722.66 A. Dsbierne ( 1 ) Ann. physiybe, 1914, [ixJ, 2, 97 ; A . , 1915, ii, 126 ; (2)ibid., 425, 478 ; A., 1915, ii, 132, 725, 726 ; (3) ibid., 1915, [ixl, 3, 18, 62;A . , 1915, ii, 302, 303 ; (4) ibid., 1915, [ixl, 4, 408 ; A., 1915, ii, 667.6 i R. Heimann, Monatsh., 1914, 35.1533 ; A . , 1915, ii, 665.6 8 Ann. Report, 2909, 263.69 (Mlle.) E. Gleditsch, -4mer. J . Sci., 1916, [iv!, 41, 112; A . , ii, 168266 ANNUBL REPORTS ON THE PROGRESS OF CHEMISTRY.make sure all ionium had been removed. The two most satisfactoryexperiments gave the period of half-change as 1660 years, in goodagreement with Rutherford's value, 1690 years, obtained by count-ing the a-particles expelled per second from a known weight ofradium. The value for the period of average life is 2347 years,somewhat less than the value, 2500 years, usually taken.Polonium.-Some indirect evidence has been obtained of theexistence of a volatile hydride of polonium, in the course of experi-ments on the range of the a-particles in hydrogen, in which agradual increase in the ionisation current a t a given distance withthe lapse of time was observed.70 Being in the sixth family ofelements, it is to be expected t h a t it should form a volatile com-pound with hydrogen dissociating a t the ordinary temperature.The Emanations.-A lengthy series of experiments has beencarried out on the volatility of the thorium and radium emanationsto see if these two isotopes, once mixed, could be separated by con-densation a t low temperature.71 A large number of puzzling pheno-mena were encountered, and the results obtained varied widelyfrom experiment to experiment.Owing to this, and the greatdifference in the periods of average life, and, consequently, in theconcentration of the two emanations, no definite conclusions couldbe reached on the main question, which remains unanswered.I nthe course of the investigation, two maxima were observed incertain cases in the condensation curve of radium emanation, onea t - 7 5 O and the other a t - 1 6 1 O . This may be connected withthe well-known phenomenon that, when a tube containing emana-tion is immersed in liquid air, the condensation, as shown by theluminous ring on the tube, occurs always a short distance abovethe level of the surface of the liquid air.The Branching Poiti t of t h e Thorizcm Series.72-The supposedseparation of the two sets of thorium-C atoms, which give a-rays ofrange 5.7 and 8.16 cm. respectively, has not been confirmed. Thetwo stages in the volatilisation, 35 per cent. being volatilised belowand 65 per cent.above 900°, has been confirmed, but it has beenshown t h a t the isotope radium4 shows in every respect a com-pletely identical behaviour. This a t once rules out the explanationthat the 35 per cent. of the thorium-C atoms which give short-range a-rays had been separated from the 65 per cent. which givea-rays of long range, because in this case, although radium-C dis-integrates dually also, all but an infinitesimal fraction of the atomsio R. W. Lawson, Monatsh., 1915, 36. 846 ; A . , ii, 121.72 -4nn Report, 1914, 287; S. JJoriit, Ph,ysikal. Zeitsch., 1916,17, 6 ; Monatsh.,A. Fleck, Phil. Mac/., 1916, [vi], 29, 337 ; A . , 1915, ii, 131.1916, 37, 1 7 3 ; A . , ii, 169, 465RADIOACTIVITY. 267follow one of the two modes.These C-members are isotopic withbismuth, and it is probable that, like bismuth, they form severaloxides stable within different limits of temperature, which accountsfor the discontinuity in the volatilisation curve.Adsorptiou of the Radbelenien ts.The definite chemical characterisation of all the known radio-elements raises the problem as to how it' is they show such definitebehaviour in so dilute a solution, and are precipitated, along withother precipitates, a t ionic concentrations far below the solubilityproduct. Several papers have been published on this subject.73M7here the substance carrying down the radio-element is isotopicwith it, absorption, of course, can play no special part, for theratio of t h s concentrations of the two isotopes in both the solidand liquid phases must be the same, and the ratio of the quantityprecipitated to the quantity left in solution also the same for each.Where, however, the two substances are not isotopic, but merelyanalogous, the radio-element is carried down by the precipitate ifunder the same conditions it would be precipitated if present insufficient concentration. The various workers agree on the con-clusion that the negative ion governs the precipitation.Adsorp-tion is favoured if the adsorbent has an electronegative constituent,the compound of which with the radio-element is insoluble.Similarly, the addition of an acid containing such an electronegativeconstituent favours the adsorption. Adsorption in these cases isquite in keeping with the chemical character, whereas in the caseof some of the adsorbents firstl used, such as charcoal, entirelydifferent considerations may apply. An exhaustive study of theadsorption of uranum-X, by the last-mentioned substance hasshown that the action of the isotope, thorium, in preventing theadsorption, discovered by Ritzel, is shown also by a large numberof substances-zirconium salts, benzoic acid, strychnine nitrate,and basic dyes.74 It was found that a solution of thorium nitratewhich has been shaken with charcoal produced afterwards a muchsmaller effect in preventing the adsorption of uranium-X1 by char-coal.Similarly, uranium-X1, freshly produced from uranyl nitratet h a t has been shaken with charcoal, is more readily adsorbed bycharcoal.The authors assume the existence of small quantitiesof still undiscovered radio-elements in thorium and uranyl nitrateswhich are the cause of the effect, and are removed when shaken73 I<. Horovitz and F. Paneth, Zeitsch. physikal. C'hem.. 1915, 89. 513 ;&4., 1915, ii, 215 ; F. Pnneth, Physikal. Zeitsch., 1914, 15, 924 ; A . , 1915, ii,2 0 5 ; K. Fajans and F. Richter, B e y . , 1915, 48. 700; A . , 1915, ii, 406.n H. Freundlich and H. Kaempfer, Zeitsch. physikal. Chein., 1915, 90,681 ; A . , ii, 70268 ANNUAL REPORTS Oh’ THE PROGRESS OF CHEMISTRY.with charcoal. Thorium nitrate, unlike the other substancesinhibiting the adsorption, is eRective if added to the charcoal afteri t has adsorbed uranium-X1, causing the latter to redissolve. Itwas recognised long ago’s that, if thorium nitrate did not preventthe adsorption of the isotope uranium-S1, this means would beavailable for separating two isotopes, assuming, of course, that thethorium nitrate itself was not, equally with uranium-S,, adsorbedby the charcoal.The fact that other substances have the sameeffect as thorium in no way affects this coiiclusion, but the diminu-tion of the effect of thorium nitrate by shaking the solution withcharcoal requires an explanation if these authors’ somewhat sweep-ing assumptions of undiscovered radio-elements is not accepted.Perhaps, t o hazard a suggestion, the thorium nitrate itself is moreeasily absorbed by charcoal after its solution has been shaken withcharcoal, which naturally would make it less effective in preventingadsorption of uranium-X,.The method of adsorbing radium by colloidal silicic acid, with aview to the subsequent volatilisation of the latter by hydrogenfluoride, has been found unsatisfactory in practice, being verysensitive t o the presence of acids and t o variations in the characterof the silicic acid gel.76 Manganese dioxide hydrate, preparedeither by reducing a permanganate with methyl alcohol or withmanganese chloride, has been found suitable, and here it is to benoted the action is chemical, depending on the formation of aradium manganite.An enrichment of the radium from bariummay be effected by partial de-adsorption by the electric current,but the best method is to treat the dioxide hydrate with aluininiunchloride solution (15 grams of crystallised salt per litre), whichreplaces more radium than barium in the manganite.It isdoubtful from the figures given whether the process is so simpleand effective as Mme. Curie’s original method of fractionallycrystallising the chlorides.T e c J w z icnl T1-m t I ~ L e I I t of R cr tl ioci c t i f 6 M(i t e 1.ictl.7.The same authors recommend for the reduction of crude radiumresidues, consisting mainly of lead and alkaline-earth sulphates, tosulphides soluble in acids, a mixture of calcium hydride andcalcium carbide, the latter moderating the violent action of theformer.77 A proportion of one to three suffices for rich residues,75 Ann. Report, 1910, 276.T G Ibid., 1911, 292 ; E.Ehler and W. Bender, Zeytscli. nngezc. Chem.. 191,5?7 7 Compare -4~2n. Report, 1913, 377 ; E. EhIer and W. Bender, Zeitsch.28, 35, 41 ; A . , 1913? ii, 659; 1915, ii, 129.anorg. Chejn., 1914, S$, 25.5; A , , 1915, ii, 404RADIOACTlVITY. 260but more hydride must be added for the poorer varieties. Forthe separation and concentration of radium and its isotopes frombarium, the fractional precipitation of barium hydroxide from itssolution by addition of an alkali hydroxide has been patented.78Ionium and actinium have been recovered, from the crudesulphates of the Olary ores, from the filtrate from which thealkaline earths have been separated by sulphuric acid.79C'ctrnotite.-The report of the Bureau of Mines, Washington,Bulletin 104, Mineral Technology 12, dated November, 1915, onthe extraction and recovery of radium, uranium, and vanadiumfrom carnotite, contains an interesting account of the steps takenin America to nationalise the extraction of radium.I n an experi-mental plant, from which 4.25 grams of radium (element) wereextracted, the cost worked out a t 37.6 dollars per milligram, ofwhich 20.71 dollars was the cost of the extraction and the rest thecost of ore. This is on the basis that the uranium and vanadiumwere not recovered. Actually they were recovered, and it isexpected will more than pay the cost of recovery. Full detailsare given of the factory operations and methods employed, andthe report would be invaluable to anyone wishing to start extract-ing radium technically.A method which seems promising, previously used in the estima-tion of radium by the emanation method,80 has been applied tothe extraction of radium from carnotite.81 Boiling a low-gradeore with 96 per cent.sulphuric acid removed 95 per cent., 78 percent. sulphuric acid 92 per cent., and 35 per cent. sulphuric acid42 per cent. of the contained radium, the latter strength sufficingto remove the vanadium and uranium effectively. I n actual workwith 10 kilogram lots, 18 kilos. of acid, 60° Baum6, containing78 per cent. of acid, were heated t o 190°, and the ore addedgradually, with stirring, heating being continued for fifteenminutes a t least until the temperature reached 220O. The mixturewas filtered on a " Filtros '' medium, washed with two lots of freshhot acid, and the filtrate run into eight times its volume of hotwater and well stirred.The radium is precipitated with thebarium as sulphate, and in this one operation its concentration isincreased more than 230 times. Several per cent. of the radiumsettles out from the turbid filtrate obtained by washing the residue7 8 H. N. McCoy, U.S. Pat., 1103E,OO, J . SOC. Chem. Ind., 1914, 33, 919;7 9 S. Radcliff, J . Roy. SOC. New 80zith Wales, 1914, 48, 408; A . , 1915,R o -4m. Report, 1911, 294.A . , 1915, ii, 3 .ii, 665.TT. Srhliiurlt, J . Physicnl Chern., 1916, 20, 485 ; A , , ii, 430270 ANNUAL REPORTS ON THE PROGRESS O F CHEMISTRY.on the filter with water. An actual recovery of 85 t o 95 percent. of the radium is recorded in four experiments.A careful investigation of various methods of treating carnoti te,noteworthy because all the radio-constituents were traced by a-raymeasurements, showed a recovery of 90 per cent.of the uraniuniand of the radium, 50 per cent. of the polonium, 61 per cent. ofthe ionium, and 52 per cent. of the actinium.82 The treatmentinvolved boiling the ore (1) with sodium carbonate to remove theuranium, which separated in a pure form as sodium uranylcarbonate on concentrating the filtered solution ; (2) with hydro-chloric acid, which removed most of the radium; ( 3 ) with nitricacid, some 8 to 10 per cent. more of the radium being obtained;(4) with moderately concentrated sulphuric acid, which removedthe ionium.I t is interesting that the ionium was not removedin the first three treatments. The residue, which was reduced inweight to one-half t h a t of the mineral, still contained 4.2 per cent.of the initial radium. The carnotite used was not pure, but con-tained much of its vanadium in a very difficultly attacked form,probably as the silicate, roscoelite. Only part of the vanadiumwas recovered even by heating the ore in dry hydrogen chloride,when the volatile oxychloride distils away. This method is statedto remove the whole of the vanadium from the crude sodiumuranate obtained as a by-product from cariiotite.83 Other methodsof effecting the latter separation are (1) the heating of a paste ofthe uranate and ammonium chloride with water; (2) solution ofthe uranate in the minimum of dilute acid and boiling, when thevanadium and 13 per cent.of the uranium are precipitated.Mndapacczr Mitz emT.y.-The districts of Antsirabe and Betafo,Central Madagascar, 100 miles S.W. of Tananarive, is stated t o beextraordinarily rich in minerals containing some 20 per cent. ofuranium along with columbium, tantalum, and titanium. Thecomplete analyses of four, blomstrandite, betafite, samiresite,ampangabeite, by Lacroix, show 18.1, 26.6, 21.2, and 19.4 per cent.of UO, respectively, and also in the second 1.3 and the fourth2.5 per cent. of thorium oxide.8' I n a new examination of pyro-morphite from various sources by radioactive methods,sS the surf aceconcentration of the radium is denied, and the mineral is statedt o be homogeneous as regards its radium content.It is regardedas a young formation, in which radium, deposited a t the time of83 H. 31. Plum, J . r1?)29T. Che7n. SOC , 1315, 37, 1797 : L4., 1915, ii, 666.83 H. 13. Rarker and R. Schlundt, J . POC. C'hcm. Ind., 1916, 35, 176 ; -4.,84 T. P. Waites, J . Chem. Met. Min. SOC. S. A4.frica, March, 191C, p. 187.85 M. Bamberger and G. Weissenbtrger. Monntsh., 1915, 36, 169 ; A.,i, 189.1915, ii, 506 ; compare Ann. Reports, 1909, 260 ; 1910, 264R AD1 0 ACT I\' I TY. 271its formation from the water in which it was formed, has not yethad time to decay.A r ti fic ial Tra 11 smut a tio n .In aseries of electric discharge experiments in hydrogen with differentsized coils, different types of interrupters, various sized and shapedtubes, with palladium, platinum, and aluminium electrodes, noproduction of helium or neon was observed,8$ which confirms theview previously taken in these Reports that these gases are notobtained when due precautions against contamination are taken.Uranium oxide subjected t o cathode rays lost oxygen t o someextent, but there was no evidence t h a t the compound had beenrendered more radioactive.Bismuth, similarly treated, was notrendered active, and although it showed subsequently a spectro-scopic trace of thallium, this was present also in the untreatedmeta1.87 Lastly, attempts made to influence the velocity of radio-active transformation by a-rays were unsuccessful, uranium oxideand mesothorium preparations being exposed t o the bombardmentof the a-rays from the radium emanation without any change intheir radioactivity, as subsequently measured, being produced.**The results to be recorded in this field are all negative.iQa t ural R udioa c t i L' it y .A radioactive determination of the thorium content of 86 acid,48 intermediate, and 56 basic rocks gave mean values of 2.1, 1.5,and 0.5 ( x 10-5 gram per gram), with a general mean of 1-4.89Analyses of the ochre deposited by a strongly radioactive Tyrdlspring showed a concentration of radium several times greater thanthat in the rock from which the water issues, which was a graphite-quartzite containing zircon and about 0.1 per cent. of t h o r i ~ n i .~ ~A connexion between fertility of the soil and radioactive contentwas found in the case of thirteen typical Minnesota soils, thosericher in radioactive constituents being without exception the morefertile.91A comprehensive survey of the radioactivity of 400 Swedishspring waters showed a relatively high radium content, the most86 A. C. G. Egerton, Proc. Roy. SOC., 1915, [A], 91, 280; A., 1915, ii, 132.87 W. P. Jorissen and J . -4. Vollgraff, Zeit.sch. physikal. Chem., 1914, 89,151 ; 1915, 90, 557; A., 1915, ii, 134 ; 1916, ii, 71 ; also A., 1915, ii, 664.J. Danysz and L. Wertcnstein, Gompt. rend., 1915, 161, 784 ; A., ii, 69.J. H. J. Poole, Phil. May., 1915, [vi], 29, 483 ; A . , 1915, ii, 207.G. Weissenherger, C'e?z.fr. Min., 1914, 481 ; A . , 1915, ii, 305.91 J. C. Sanderson, Amer. J. Sci., 1915, [iv!, 39, 391 ; A . , 1915, ii, 306272 ANNUAL REPORTS ON THE PROGRESS O F CHEMISTRY.radioactive-up to 172 Mache units-being those from deepborings in the acid rocks-granites and syenites.91 Two springs,the first in Bloomington, Indiana, and the second a t Tuwa, Kaira,Bombay, in which the emanation content varied with the seasonalvariation of the flow of the spring, showed precisely oppositebehaviour. I n the first case the emanation content increased, andin the second case decreased as the flow of the spring increased,suggesting in the latter case a constant supply of total emanationall the year round, independent of flow, and in the former a supplyderived from the surrounding soil and depending on the rate ofpercolation of the water, less decay en route taking place the morerapid the percolation and greater the rainfall.93Obit L l c r T y .I n conclusion, it is fitting to recall the great losses the sciencehas suffered in the period under review by the deaths of €1. G. J.Moseley, in his twenty-eighth year, in the fighting a t Suvla Bay,Gallipoli, the youngest investigator, surely, to have won so securea place in the history of science, and of Sir William Ramsay, who,pre-eminent in chemical science before the discovery of radio-activity, devoted the last ten or twelve years of his vigorous andcrowded life more and more exclusively t o the young science he didso much to advance.FREDERICK SODDY.s3 N. Sahlhom, Srkiv Kern. Min. Geol., 1916, 6, No. 3, 1 ; A . , i i , 208.g3 R. R. Ramsay, Phil. Mag., 1915, [vi], 30, 815 ; A., ii, 6 ; A. Steichen,ihid., 31, 401 ; 9., ii, 28-1.ADDENDUM TO FOOTNOTE 4, PAGE 247.An interesting and important confirmation has just' been receivedthrough Dr. Lawson, who is interned in Vienna and allowed towork in the Radium Institut under Professor Stefan Meyer. Thewriter sent Dr. Lawson the first fraction of the above distilled thoritelead, and he now reports (January 31st, 1917) that ProfessorHonigschmid has made four determinations of its atomic weightby the silver method, the mean result found being 207.77 k0.014,in excellent agreement with the value 207.74 calculated from therelative density of the specimen as found by the writer.F. 8

ISSN:0365-6217    

DOI:10.1039/AR9161300245

出版商:RSC    

年代:1916

数据来源: RSC

摘要:

iNDEX OF AUTHORS’ NAMES.Abderhalden, E., 74, 96, 205, 208.Abelmann, A., 175.Ackroytl, H., 210.-4dams, E. Q., 30.Adanti, G., 186.Addis, T., 216.Addy, C. W., 156.Aguilar, R. I€., 192.Alessandri, L., 141.Alqar, J., 151...411en, E. R., 187.Allen, E. T., 54.Allison, F. E., 234.Almstrom, G. K., 140.Alway, F. J., 227.Amberger, C., 43,Ames, J. W., 342.Andersen, A. C., 205.Anderson, E., 85.Anderson, G., 173.Anderson, R. P., 167, 168.Angeli, A,, 122, 141.Arndt, K., 58.Artis, R., 220.Aschsn, O., 117, 118Atterberg, A., 226.Aubry. A., 89.Aust, E., 157.Auwers, K. voii, 118, 149, 150.Averkiev, N., 65.Backer, H. J., 190.Bacon, W., 185.Baeyer. O . , 262.Bailly, O., 74.Raker, J. L., 179.Balke, C. W., 2-7.Baly, E. C. C., 6, 7, 8.namherger, M., 270.Rang, I., 212, 213, 216.Rargellini, G., 100.Barger, G., 162.Barker, H.H., 60, 270.Barkla, C. G., 3.Barnehey, 0. L., 185.REP.-VOL. XIII.Barnes, J., 259.Barratt, J. C. W., 200.Baudisch, O., 125.Bauer, H., 223.Raster, G. P., 33, 247.Beal, G. D., 181.Beans, 13. T., 21.Beattie, J. H., 230.Beck, R. P., 40.Bender, W., 268.Bennett, A. H., 189.Bennett, C. W., 47.Renton, T. H., 228.Berg, J. C. van den, 92.Rergmann, &I., 75, 86.Rerman, A. C., 55.Berkeley, Earl of, 19.Bevan, E. J., 165.Beyerinck, M. W., 92.Biederman, W., 167.Biehringer, J., 62.Bingham, E. C., 166.Bishop, E. S., 227.Bissel, D. W., 50.Bizzell, J. A,, 223.Blanchetk+e, A., 175.Blish, M. J., 204, 227.Blumenfeld, J., 45.Bodenstein, If., 64.Bodinus, F., 154.Boeseken, J., 87, 92.Boehm, R., 157.Bogert, M.T., 105, 139.Bohr, N., 252.Bokhorst, S. C., 40.Boltz, G. E., 242.Bonardi, J. P., 49.Booge, J. E., 17.Rorsum, W., 62.Bose, D., 261.Bournot, K.. 157.Rourquelot, E., 89.Bradley, H. C., 207, 208.Brady, 0. L., 126.Brady, St. E., 181.Bramley, A., 16274 lNDEX OF AUTHORS’ NAMES.Brannigan, 1’. J., 8.Braun, J. von, 101, 102, 138,Rrauner, B., 185.Brauns, D. H., 86, S7.Brazier, S. A,, 77.Breteau, P. 56.Bridgman, P. W., 30.Brieger, W., 77.Brill, H. C., 174.Briner, E., 59.Broek, A. van den, 3, 254.Broglie, M. dt:, 259, 260.Bronfenbrenner, J., 203.Brown, D. J., 190.Brown, P. E., 232, 234, 236, 237.Browning, P E., 52.Bruhns, G., 175, 184, 191.nruni, C., 53, 64.Brunner, E., 100.Buckley, J.P., jun., 180.Ruckner, G. D., 212.Bull, H., 180.Burgess, G. K., 38.Burgess, P. S., 235.Burrell, G. A., 169.Burt, F. P., 31.Busch, &I., 93, 100, 125.Cain, J. R., 188.Callow, R. H., 8.Campbell, F. H., 14.Cardoso, E., 11.Carter, E. G., 233.Casale, L., 124.Castets, J., 173.Cavazzi, A., 55.Celichowski, 171.Challenger, F., 128.Chancel, F., 190.Chaplin, R. M., 185.Charrier, G., 124.Chattaway, F. D., 94.Chou, T. Q., 155, 159.Claasz, M., 147, 145.Clayton, A., 80.Cleaves, H. E., 188.Clibbens, D. A., 97.Cohen, E., 38.Cohen, J. B., 06.Coleman, A. B., 166.Coleman, D. A., 231.Conant, J. U., 190.Conn, E. J., 232.Conner, S. D., 225.Cook, F.C., 242.Cook, K. C., 231.Corell, M., 162.Cox, A. W., 166.Cramer, F., 64.Crawford, F. M., 190.C r i s p , D., 171.Crowther, C., 202.157, 158.154,hrtis, I<. E., 232.7urtiu=; T., 132.h r t m a n , L. J., lS2.jaish, A. J., 239, 240.>akin, H. D., 96.Pale, H. H., 198, 202.M e , J. K., 86.janysz, J., 271.jarapsky, A., 05, 132.)aschavsky, P., 182.l a t t a , A. K., 63.Javis, W. A., 172, 178, 239, 240.Iavisson, R. S., 187.Dean, H. I?., 113.Debierne, A., 257, 265.Jeniges, G., 136, 173.Dennis, L. M., 46, 47.Derick, C. G., 113.Uevaux, H., 241.Dhar, N., 63.Diakonoff, (Mlle.) 204.Dickson, T. W., 26.Uiesselhorst, H., 22, 42.Dimmitt, F. W., 217.Dimroth, O., 114, 115.Dittlor, E., 180.Dodge, F.D., 148.Dorka, C., 73, 119.Dubsky, J. V., 95.Dufresne, M., 96.Dunlop, (Miss) J. G., 3.Dunn, F. P., 126.Eastlack, H. E., 21, 51, 52.Ebler, E., 268.Edgar, E. C., 31.Edgar, G., 188.Edson, It., 59.Edwards, A., 177.Edwards, V. C., 107.Egan, J. E., 47.Egerton, A. C. G., 271.Egloff, G., 68.Ehrenberg, P., 243.Ehrlich, J., 176.Eichwald, E., 74.Elbs, K., 100.Elias, A., 139, 146, 149.Ellenberger, W., 232.Ellingworth, S., 24.Emmert, B., 155.Engle, W. D., 186.Epps, G. D. van, 83.Fajans, K., 1, 252, 267.Falk, K. G., 187.Fazi, R. de, 112, 173.Felsenreich, G., 119.Ferman, A., 265.Fichter, F., 100INDEX OF AUrHORS’ NAMES. 275Fick, R., 114.Field (Miss) E., 162.Fine, 11. S., 217.Fink, H., 75, 140.Fischer, E., i 5 , 77, 84, 86.Fleck, A., 266.Fleischer, K..112, 160.Fodor, A., 06.Footc, H. W., 41.Foreman, F. W., 210.Found, C. G., 260.Frnncli, J., 5.Franqois, &I., 173.Frankel, E. RI., 206.Franklin, E. C., 220.Frary, F. C., 55.li’razer, J. C. W., 18.Freurrd, JI., 112, 159, 160.Fremxcllich, H., 20, 22, 23, 42, 267.Friedlaender, P., 147, 154.Friedmann, W. , 145.Friend, J. A. N., 133.Frimnn, E., 257, 258.F ~ i r t h , 0. von, 119.Fnnii, C., 206.Furtnan, N. H., 190.Gabricl, S., 147.Gadanier, J., 157.Gainey, P. L., 233.Garnier, C., 45.Garrett, C. S., 6, 7.Gattermann, L., 154.Geake, A., 94.GBrard, 33.Gerhardt, &I., 1 16.Cermann, F. E. E., 36, 170.Ghosh, €3. N., 150.Gibbs, W. E., 233.(:illespie, L.J., 225.(:leditsch, (Mllc.) E., 26.5.(iluud, W., 153.(ioldenzweig, 149.(jolse, J., 192.Ciornberg, M., 67.GonzBlc~; A.; 181.Goost, T., 03.Gortner. R. A., 172, 204, 205, 228.Gould, R. A., 220.Graffenried, A. von, 189.Graham, G., 214, 218.Gramont,, A. de, 37.Grant, A. J., 50.Greaves, J. E., 233, 235, 237.Grignard, V., 17.5.Grimmer, W., 232.Qrindley, H. S., 204, 205.Grose, M. R., 33.Grover, F. L., 21-7.Griinkraut, A., 45.Griinzweig, M., 44.Griittner, G., 127, 124, 131.Gsell, H., 167.Gustavson, R. G., 186.Guye, P. A,, 36, 170.ITaagen, W. K. van: 31.Hager, G., 178.Hahn, O., 262.Hall, R. E., 2, 3.Hamilton, (Miss) E. R. P., 79.Hantzsch, A., 75, 107, 108.Harding, T. S., 90.Harding, T’. J., 95, 173.Harkins, W.D., 2, 3, 253.Harries, C., 73,.Harrison, J. AT., 94.Harrison, W., 92.Hart, E. B., 242.Hartley, E. C. J., 19.Hartley, P., 203.Hartman, 31. L., 183.Hartmann, M. L., 3 3 .Hartwagner, F., 186.Hartwich, F., 156.Hatschek, E., 22.Haworth, W. N., 90.Headden, W. Y., 243.Heimann, R., 265.Heise. G. ?IT., 192.Hekma, E., 201.Helderman, W. D., 38.Hertz, G., 5.Herzfeld, E., 306.Herzop, W., 147.Hess, I<., 75, 140, 141, 143, 163.Hevesy, G. von, 1.Hildebrand, J. H., 15.Hildehrandt, F. M., 223.Hilgard, E. W., 220.Hillers, D., 95.tlinsberp, O., 115.Hitchins, (Miss) A. F. R., 249.Knagland. D. R., 226.Hoeflake. (Mlle.) J . M. A , 103.Honigschmidt, O., 34,35, 517, 248. 230.Hoffmann, F., 169.Hoffmann, La Roche & Co., F., 159.Hofmann, K.A., 72.Holden, H. C., 48, 49.Holland, E. R., 180.Holmberg, B., 78.Holmes. J. A, 220.Holtz, J., 11 I, 137.Hopffc, A,, 232.Hopkins, F. C., 210.Horovitz, K., 267,Horovitz, (hllle.) S., 34, 3 S , 247, 2-28,Howell, W. H., 198, 201.Hudson, C. S., 84, 86, 87, 90.Hiihncr, R., 73, 97.Hiittner, C., 169.Hulett, G. A., 33.250.I, 276 INDEX OF ATJTHOBS’ NAMES.Hulton, H. F. E., 179.Hunt, C. H., 171.Hymaii, H., 36.Ikeuti, H., 261.Trrine> J. C., 85, 87.Tselin, If., 118.Issoglio, G., 1Y0.Jaeger, F. M., 4.1.James, C.. 47, 48, 49, 50, 54.Jamiesoii, C:. S., 1Sci.Jaiisen, H. C. P., 217.Jnntsch, G., 45.cTaquerod, A . , 63.Jenner, F. W., 110.Jessen, W., 116.Johansson, S., 226.Johnsnn, H.W., 237.Johnson, J. XI., 84, SO.Johnson, L. C., 33.Johnston, 17. S., 223.Johnstone, J., 55.Jollee, A., 147.Jonn, T., 129.Jones, F. H., 177.Jones. H. C., 7.Joriswn, ITr. P., 271.Jiilicher, C., 148.Kaempier. H., 267.Kailaii, A., 265.Kam, J., 11.Kamm, O., 113.Knppen, H., 224.Kastle, J. H., 212.Kaufmann, A., 156, 157.Kay, S. A., 193, 194.Kecsom, W. H.. 3s.Krhrmann, F., 148.Kelber, C., 69, 100.Kellherg, T. N., 3s.Kelley, G. L., 190.Kendall, J., 17.Kenyon, J., 96.Kern, J., 178.Kerstjens, A. H., 87, 92.Kinoshita, S.. 261.Kishrwr, X., 116, 143. 154.Klamer, C. E., 87.Klein, 11. .4., 235.Kiiapmaii, F. G. W., 170.Kohler, E. F., 102.Kopeloff, N.. 231.Koschclev, F. F., 70.Kragen, S., 185.Krall, I-I., 9,5.Krausc, E., 128, 131.Kremann, R., 15.Kriiyt, H.PI., 23, 42.Kuiider, H.. 125.Kiiiiz-Krausc, H., 72.Law, J. J . \-an, 11, 12.La Forge, F. B., 88.Laininan, (Miss) E. H., 30.Lnnp, J., 5 .Lantsbcrry, W. C‘., 261.Lapworth, A , , 82.Lathrop, K., 235.Lallg, -1., 58.rdaiv, .J., w.Lnu son, R. w., 206.rJelllOl1, R. J., 46.r,O R ~ ~ , G. A., 192.Ledrrer. K., 128, 120.Lciiher, V., 187.Lerch, H., 100.Levrne, P. A., 86, 88, 89, 96, 200.Lcvi, Q., 53, 64.Lewis, G. N., 30.Lewis, W. C . M., 8, 28.Lifschiitz, I., 110.Linderrnaim, F. ,I., 752, 237.Lint, H. C., 231, 237.Lipinan, C. U., 221, 227, 243.Lipman, J. C.. 237.Lockcrnaiin, (;., 264.Lob, ITr., YO.Lohnis, F., 230.Lomhroso, U., 214.Longtnnn, J . , 9 7 .Lorin, S., 266.Lough, U T .G., 817.Lowry, ?‘. M., 26.Luff, B. D. W., 71.Lutz, 0. E., 78.Lyoii, T. L., 223.Macheth, A. K., 8, 150.McBeth, J. G., 232.McCall. A. (i., 823.McCay, L. W., 190.RIeCliigage, H. 23., 11 3.MeCoy, 1-1. N., 269.Macdoiiuld, J . L. -\., 85.RiIacThiticll, C. C’. , 1 SX.~7scl>ougall, E’. H., !I.Alcllhilwy, P. C‘., lW.w i n t o s i i , r j . , x).XcKie, (Miss) P. I-., 176.McLean, H., 215.ILZcLean, H. C., 237.McLeaii, J., 200.MacLeari, R. M., 95, 177.McLeiman, J., 260, 261.Madinaveitia, il., 181.Rlarz, S., 223.Maillard, L. C., 96.Majima, R., 105.Maliowcr, W., 260, 261, 763, 264.Rlalmar, l . , 259.Blandel, J. A., 121.Manncssitr, A., 146.Mannich, C., 73.R;laqllt~illlc, L., 178, 179INDEX OF AUTHORS’ NAMES.Marden, J.W., 1SO.Marin, A., 244.Marsden, E., 261.Martin, E., 57.Mason;, G., 243.Mathews, A. P., 12.Mayer, M., 103.Meitner, L., 248, 262.Mellor, 13. S., 82.Meloclie. C. C., 59.Menrlel, L. B., 211.Mercer, H. V., 261.Merck, F., 141.Merton, T. R., 4, 248.Merwin, H. E., 54, 55.Meiiiiier, J., 183.Meytr, G., 40.Moyrr, G. M., 86.Meyer, H., 97.hIoycr, S . , 251.Middletoil, A. R., 1S2.Miller, H. L., 182.Minges, U. A., 236.Minot, C:. R., 199.Mitchell, H. H., 210, 211.Miyake, K., 243.Morner, C. T., 96.Moles, E., 32.Monnier, A., 184.lCIontniollin, G. de, 67.Morgan, J. C.. 48.Alorise, M., 20B, 20MUller. E., 132.illueller, J. H., 181Myers, V. C., 217.Mylius, F., 169.Myrick, y\. T., 18.Nametkin, S.S., 1Nelson. J. 31.. 92.s.Neubwg, C., 121.Xeumnnn, L.. 154.Newlands, S. H., 193, 104.N;cholson, J. W., 4, 253, 253.Nierenstein, M., 94, 97, 113.Yodder, G., 8.Nollau, E. H.. 212.PITolte, O., 187.Sorthrop, J. H., 92.Noyes, W. A., 32.Oberfell, G. G., 160.Ochs, It., 111.Oddo, n., 163.CFkhsner de Coninck, W., 33.Okazaki, Y.. 105.Omarini, L., 124.Onnes, H. K., 38.Orange, L., 73, 119.Orton, K. J. P., 176.Osborne, T. B., 211.Ostromisslenski, I. T., 68, 69, 70.Ott, E., 78.Paneth, F., 1, 2, 251, 267.Parker, H. O., 84.Patch, R. H., 102.Patttrson, T. S., 25, 26, 81.Pauli, W., 265.Paulus, M. G., 7.Pellaton, AT., 64.Pellini, Q., 63.Penau, H.. 193.Perkin, A. G.. 101.Perkin, W. H., juii., 103.160.Perley, G. A., 54.Pcsci, L., 148.Pfeiffer, P., 108, 109, 153.Pfciffer, T., 242.Philipov, O., 115.Phipps, T. E., 190.Picket, A . , 155, 159.Piercc, G., 8-3.Pilz, F., 171.Pincas, H., 61.Pisani. F., 187.Yisarshovski, L., 65.PltZ, w., 236. 242.Piva. A, 169.Plaut, E., 105.Plum, H. M., 270.Polack. TV. C., 58.Ponzio, G., 127.Poole, J . H. J., 271.Popp, JL, 171.Posnjnk, E., 51.Potmdil, K., 177.Potter, R. S., 228.Povarriiii, C,. , 93.Powell, A. RJ.. 184.Pornis, F., 21, 24.Prandtl, TV., 183.Prescott, J. A., 172, 222.Price, T. S., 77.Puxeddu, E., 127.Pyman, F. L., 158.Quinn, E. I,., 33.Rabinovitsch. P. N., 60.Radcliff, S., 263.Radosevjit, R., 166.Raikov, P. N., 98.Raistrick, H., 202.Ralrshit, J., 176.Ramann, E., 223.Rnmsay, (Sir) W., 4.Ramsey, R.R . . 272.Randall, E. L., 176.Rrtsthig, F., GO, 61.Rasmussen, H. B., 181.Rnther, J. R . , AS.Ral-enna, C., 240.Reckleben, H., 264.Reed, J. C., 180.Reid, E. E., 67, 83.Pauly, H., 108278 lNDEX OFRcinders, W., 23.Kenard, T., 36.Resau, C., 121.Rhodes, F. H., 47.Rice, J., 28.Richards, T. W., 35, 36, 249.Richardson, H., 259, 263.Richardson, S. S., 27.Richter, F., 267.Riffc, J., 168.Rigg, G. I?., 229.Roark, R. C., 188.Robertson, G., 87.Robertson, G. S., 170.Robertson, P. W., 174.Robinson, F., 177.Robinson, (Mrs.) G. M., 106, 110.Robinson, R., 79, 81, 110.Robinson, W. O., 178.Roschdestwensky, X., 154.Rose, W7. C., 217.ROSS, (Miss) W., S9.Rothlen, T“., 1<57.Rosw, M.L., 305.Rubio. C., 244.Ruff, O., 62.Rund, C., 75, 84, 86.Rupe, H., 118.Russell, E. J., 222.Rutherford, (Sir) E., 259, 260.Ryan, H., 151.Rydbcrg, J. R., 2.Saclis, J. H., 192, 193.Sahlborn, N., 272.Sahni, R. R., 261.Salvadori, K., 182.Samec, &I., 92Sander, A., 6 2 , 186.Sanderson. J . C., 271.Sanclyvist, H., 23.Savill, C. A., 166.Sawyer, G. C., 239.Saxton, B., 11.Sayre, R., 90.Scatchard. G., 139.Schaefcr, C., 8.Schncffer, E. J., 7.Schefftbr, F. E. C., 61.Scheunert, A., 232.Schlenk, W., 111, 129, 137.Schlcsinger, €1. I., 166.Sckilundt, H., 60, 269, 270.Schmidt, E., 164.Pchoch, E. P.. 190.Schoeller, W. R., 184.Schol!cnbcrger, C‘. J., 172.Schroctler, 2 2 0 .Schubert, hl., 8.Schultzc, E., 11.3.Scliultze, K., 243.Scliulze, A., 169.Scl?Oltl, 31..1 G o .AUTHOIIS’ NAMES.Schumpclt, K., 72.Schweikert, G., 38.Seligman, R., 66, 57.Senglet, R., 59.Senior, J. K., 95.Sharp, L. T.. 226.Shipley, J. W., 168.Shorter, S. A., 24.Siegbahn, M., 257, 255.Sieverts, A , , 55.Silberstein, L., 253.Simmermachcr, W.. 242.Sirnonis, H., 139, 146, 119.Singh, T3. K., 93.Skaupjr, F., 5.Skinncr, J . J., 230.Sbita, A., 73, 100, 154.Skrabnl, A., 29.Glater, 31. E., 804.Smith, A., 51, 5-3.Smith, 131. F., 31.Smith, J. W., 11.5.Smitli, L.. 2-30.Smith, N. R., 230.Smith, T. O., 48.Smits, A., 4n.Smoluchowski, n1. von, 22.Soci6ti: pour 1’Tndustrie C‘himiriue tiSodcly, 17.. 36, 247, 249, 264.Sommer, F., f i l .Spencer, S.I?., 52.Speyer, E., 159.Spiegel, I,., 162.Spicllinann, P. N., 177.Stadniliox-, (7. Td., 82.StartBk, V.. 175.Stehelin, P., 15.5.Steichen, A.. 273.Stenstrom, W., 258.Steven, 9. I., 264.Stove, H., 100.Stoklaqa, J., 241.Stubblefieltl, B. M., 187.StucLart, P., 73.Stutzer, A., 243.Suchier, A . , 163.Sugiura, K., 187.Suto, K., 97.Snarts, F., 98.Szebhyi, I’., 174.S~midt, (Miss) J., 263.Taylor, H. S., 265.Taylor, J., 208.Taylor, R. I,., 185.Thornton, W. M., jun., 18s.T’ilgner, &I., 193.Tingle, A., 183.Tinker, F., 14, 19.Totmii, G . , 210.Tottinghani, W. E., 242, 243, 244.‘L’curpnian, M., 6 3 .Bble, 159IXDEX OF AUTHORS’ NAMES. 279Truniger, E., 178.Truog, E., 187, 225.Tryhorn, F. G., 8.Tuinzing, R.W., 171.Tunstall, N., 260.Turner, E. E., 137.Turner, W. A., 189.Turski, J. F. (18, 97.Tuttle, J. B., 174.Twomey, T. J., 68.Uibrig, C., 120, 141.Urbain, G., 45. 259.Vageler, H., 243.Valori, B., 122.Vanino, L., 186.Varder, R. W., 262.Vargolici, B., 157.TTesclcy, V., 88.I’oelcker, J. A., 242, 2443.T’ollgmff, J. A., 271.Votokk, E., 88, 177.Wadsworth, C., 35, 36, 240.TF7aggaman, W. H., 244.Waites, T. P., 270.Waksman, S. A., 231, 232.Wallach, O., 116, 118.Walmsley, H. P., 263.Walpole, G. S., 198.Walsh, M. J.. 151.Walters, El. H., 228.Walther, R. von, 73, 97.mTare, E. E., 167.Warneford, F. H. S., 172.Watnnalx, C. K., 216.Waters, C. E., 174.Watson, -2. R., 150.Waynick, 13. D., 221.Weaver, E, R., 169.Wedokind, E., 93.Wcerman, R. A,, 88.Weigwt, F., 167.Weissenberger, G., 270, 271.Weitz, E., 52.Wells, T X G., 203.Wertenstein, L., 263, 271.West, C. J., 200.Westphalen, T., 119, 121.Wheeler, ,4. R., 107.Wheeler, E. G., 177.Whittemore, C. F., 48.\T7iernik, N., 127, 131.Willand, P. S., 49.Williams, P., 56, 57.Williamson, E. D., 55.Wilson, E. D., 253.Windaus, A., 1 0 , 151.Winkler, L. W., 7 2 , 131, 192, 193.Wisp, L. E., 228.Wissing, F., 143, 163.Witzemann, E. J., 88.Wbhler, L., 44.Wolffenstein, R., 156.Wolkoff, AT. I., 221.Wood, A. B., 260, 264.Wood, n. o., 19.Worley, F. P.. 14.M7riglit, Ii., 37.Wyatt, P. A., 241.TVynnc, w. l’., 220.Zappi, 33. V., 132.Zincke, T., 148.Zung, E., 204.Zweigbergk, N. von, 182.Zyl, J. P. van, 223

ISSN:0365-6217    

DOI:10.1039/AR9161300273

出版商:RSC    

年代:1916

数据来源: RSC

摘要:

INDEX OFAbsorption spec t ra .absorption.Acetal, estimation of, 176.Acetone, estimation of, 176.Acetylene, estimation of, 16:).Acids, organic arid their derivatives,S ec Spec tr a ,77.additive compounds oi, 17.Acidity of soils, 224.Additive compoiinds of organic acids,Agricultural analvsis, 170.Alcohol, estimatibn of, 173.Alcohols and their derivatives, 71, 7 3 .Aldehydes, 73.Alkali haloids, absorption spectra of, 8.Alkaloids, synthesis of, 157.Alkyl ethers of polphydric phenols,103.Allotropy, 39.Aluminates, 58.Aluminium, action of nitric acid on.17.aromatic, preparation of, 97.detection of, 173.estimation of, 181.56.hydroxides. 57.Amalgams, electrical transference in,Amines, oxidation of, 07.Amino-a,cids, 95.Ammonjatw of silver salts, 33.Ammoniurn bromide and chloride,transition temperature of, 51.Analysis, agricultural, 170.30.electrochemical, 189.gas, 167.inorganic, 182.organic, 172.physical, 1 (is.Anaphylaxis, 201.Anthracene dcrivatirc s, I 13.Antimony, compounds, 63.organic compounds, 127.Arsenic organic corn~mu~1~Xs, 132.detrction of, 182.rstirnafiori of. 187.Atmosphere, the, 219.Atom, definition of the, 251.Atoms, 1.Atom building, 252.&4toinic disintegration, cause of, 256.st rncture, 2, 262.wcights, 31.3G.crrors affectinq determinations of,Llutolysis of tissiics, 204.Azoiinide, rcactions of, 60.Azoimides, formation of, 127.Azosy-co~n~ounds, constitution of, 112.Bacteria in soils, 230.Benzol, commercial, analysis of, 177.13enzopyrones, 150.Benzoylenecarbamide, dinitro-, 139.&mzthiazolines, synthesis of, 147.Renzthiopyrones, synthesis of, 146.Uerberiiie derivatives, 160I3innry mixtures, properties of, 16.vapour pressures of, 13.Rlood, coagulation of the, 196.residual nit,rogen of the, 212.sugar of the, 214.Rromates, estimation of, 185.Bromine, atomic weight, of, 32.moIecular weights in solution of, 37.vapour pressure of, 64.detection of, 182.estimation of, 188.Cadmium, atomic weight of, 33.C a i u m , separation of rubidium and,52.Calcium, forms of, 55.carbonate, new form of, 56.hydrogen carbonate, existence of, 55.sulphide, phosphorescent, 56.28TNDEX OFCamphene, 1 17.Caoutchouc, 68.Carbides, formation of, 58.Carbohydrates, 83.analysis of.178.C’arhon, estimation o f , 174.estimation of, in steel. 188.rnonoside, estimation of, 169.Carnotite, extraction of radium,uranium, and vanadium from,269.(Jatalysis, 27.(‘atalytic reduction, 99.Cerium carbonates, 59.C’hemical reactivity, 27.Chlorates, estimation of, 183.C’hlorine, liquid, density and vapourUhlorites, 64.Cholesterol, 118.estimation of, 181.C’hromic acid. formnlil of, 63.(’hromoisornerism, 107.Coagulation of colloids, 24.Cobalt, estimation of, 186.Cobaltammines, constitution of, 133.Colloids, 41.coagulation of, 24.(‘olloidal solutions, 20.Copper, modifications of, 38.specific heat of, 38.sulphides. 34.estimation of, 188.pressure of, 64.Coumarin derivatives, 148.cl-C’rotonic acid, occurrence of, in soils,Cryptopine, 103, 160.I)ainascenine, synthesis of, 137.Dctergent action, 24.I)icyanodiarnidr~.estirnatioii o f , 177.Uifla~7niies, 151.L)isacc. h arides, 90..Uyiian~ic isomerism. 25. 106.Emtlis, rare, 43 et ?PI!.Electric discharge. 5.Elcctrical transference in amalgams,Electrochemical analysis, 189.Element, definition of the, 25 1 .Elements, chemical, heterogeneity of,high-frequency spectra of the, 257.stellar, 255.228.30.245.ISOtopIc, 1.Emulsification, 24.Enzymes, synthetical action of, 89.Equations of state, 9.Fats, aridysis of, 179.Fc: ri- and ferro-cyanides, estimationof, 176.SUBJECTS. 28 1Flavones, 130.Fluorine, estimation of, 186.Formic acid, detection of, 174.Fungi in soils, 230.Gadolinite, extraction of glucinurnGalactobiose, Fynthesis of a, 89.Gases, electric discharge tlirough, 5 .Gas analysis, 167.Glucinum, extraction of, from gado-linite, 54.Clucosides, synthesis of, 89.Glycerides, detection of, 173.Glycerol, estimation of, 180.Gold, nitrogenous compounds of, 52.Graphitic acids, 58.from, 54.specific heats of, 38.estimation of, 185.Hadlogens, estimation of, 174.Histidine, synthesis of, 158.Hydrocarbons, 67.polycyclic aromatic, 11 1.Hydrocyanic acid, detection of, 173.Hydrocyclic compounds, 11 5.Hydrogen, atomic wcight of, 3 1 .Hydroxyazo-compounds, constitutionof, 1-35!.Indcne derivatives, 11 1.Indicator, a new.139.Indigoid dyes, new class of, 147.Indigotiri and its derivatives, 154.Indole group, the, 153.Inorganic analysis, 182.Iodates and periodates, est,irnation of,Iodine, separat>ion of, 65.Todometric estimations, 185.Tonium, atomic weight of, 36.185.atomic weight, periotl, : ~ i i d sp’c-truni of, 249.Tron, estimation of, 188.Isomerism, dynamic, 25, 100.Isotopes, 1, 2.physical properties of, 253.Ketones, 73.Lead, atomic weight of, 35.atomic weight, density, and spec-orgaiiic compounds, 128.trum of, 247.Madagascar minerals, 270.Magnesium suboxide, 55.Manganese, detection of, 182.Manure, 243.Nercuric acetate as a reagent i nphosphatic. analysis of, 170.alkaloidal chemistry, 157282 INDEX OF SIJBJECY'S,Mercury, detectlion of, 182.Metals, bivalent, estimation of, 184.Mixtures, binary, properties of, 16.Molecular, structure, 2.Morphine groups, the, 155.Naphthalene derivatives, 112.Nickel, detection of, 182.Nicotine, estimation of, 181.Nitrates, detection of, 183.Nitration, 97.Nitrogen, solid, specific heat of, 38.estimation of, 186.vapour pressures off 13.weights, 37.fixation of, in soils, 237.organic, 93.compounds, aromatic, 122.reactions of, 94.estimation of, 187.salts of, 125.Nitrosoarylhydroxylamines, complexNitrosohydraeones, constitution of, 124,Nitrosulphonic acid, 62.Nitrosylsulphuric acid, 62Norcaralydine, 159.Nutrition, amino-acids in, 208.of plants, 241.Oils, analysis of, 179.Organic analysis, 172.Organo-met&ic compounds, 127.Osmotic pressure, 18.Oxidation, 100.Oximes, isomerism of, 126.Paracetaldehyde, estimation of, 176.cycloParaffins, conversion of pyr-azoline dcrivatives into, 143.N-Particles, 261.Pentazole compounds, the so-callcd,Periodic law, the, 254.Phenanthrene derivatives, 11 5 .Phenols, polyhydric, aIkyl ethers of,103.Phenylhydrazine as a reducing agent,127.Phenylhydrazines, substituted, estima-tion of, 177.Phenylhydroxylamine, nitroso-, am-monium salt, use of, in analysis,188.Phosphorus, allotropic modicaf ions of,3 9.organic compounds, 13 1.Physical analysis, 165.Phloroglucinol, estimation of, 177.Picric acid, detection of, 17:3.Plant, growing, chemistry of the, 230.Polonium, hydride of, 266.132.nutrition, 241.stimulants, ,041.Polysacchsricles, 92.Potassium dichromate, use of, i n volu-ferri- arid ferro-cyanicies, constitu -permanganate, oxidation hy, 1 S-t.Proteins, 96.hydrolysis of, 204.Pro topine, 160, 16 1.Pyrazoline derivatives, conversion of,Pyridine group, the, 154.Pyrrolc group, the, 140.metric analvsis, 183.tion of, 135.into cyrZopara!Ens, 113.Qiiiuoline group, the, 156.Radiations, chemical action of, 263.a-, B - , and y-Radiations, 260.Radioactive materials, tschnicd treat -ment, of, 268.recoil, 263.Radioactivity, natural, 27 1.Radio-elements, 265.adsorption of the, 267.Radio-lead, atomic weight of, 33.density of, 36.Radium, lifc-period of, 265.cmanation, 266.U-RSYS, 260.P-Rays, 262.7-Rays, 262.Reactivity, chemical, 27.Reduction, 98.catalytic: 99.with phcnylhydrazjne, 127.Rend permeability, 215.Rcsorcinol, estimation of, 177.Rotation dispersion.25.Rubidium, separation of calcium a ~ i d ,52.S a1 t s, complex , 1 ic: teroc y cli c tli cor i c sXcopoline, 163.Selenium, detection of, 183.Semi -permeable membranes, s true tureSilica, estimation of, 157.Silver salts, ammoniates of, 53.Sodium perborate. preparation of, 5s.of, 133.of, 19.estimation of, 190.tctrathionate, estimation of a miu-ture of sodium trithioiiate and, 63.trithionatc, action of, on mercuricestimation of a mixture of sodiumchloride, 62.tetrathionate and, 63.organic compounds, 129.Soil acidity, 22-1.bacteria, 230.fungi, 230.toxins, 229INDEX OF SUBJECTS.283Soils, 220.Solutions, colloidal, 20.Specific heats, 38.Spectra, absorption, 6.organic matter of, 226.of alkali haloids, 8.emission, 4.high-frequency, of the elements, 257.Stearic acid, estimation of, in fattyStellar elements, 255.Steric hindrance, 138.Stimulants of plants, 241.Sucrose, constit,ution of, 90.Sulphofication in soils, 237.Sulphur compounds, cyc!ic, 145.estimation of, in coal gas, 169.acids, 180.influence, 101.Telluric acid, ortho-, methyl ether of,Tellurium organic compounds, 129.Terpenes, 11 5.Thebaino, constitution of, 159.Thiazole derivatives, synthesis of, 147.Thiobenzthiopyrones, synthesis of, 146.Thiosulphates, estimation of, 186,Thorium, atomic weight of, 34.63.life-period of, 255.estimation of, 188.emanation, 266.series of radio-elements, branchingpoint of the, 266.Tin, estimation of, 186.Tissues, autolysis of, 204.Titanium trichloride in volumetricToxins of soil, 229.Transmutat.ion, artificial, of elements,Triarylmethyl, 11 1.Tungsten, detection of, 183.analysis, 184.271.estimation of, 189.Uranium, atomic weight of, 34.Vanadium, preparation of, 59.Vapour pressures of binary misturcs,T'iscosity, determination of, 165.estimation of, 189.separation of, 60.13.Water, absorption spectra of, 7.Weights, atomic. See Atomic weights.analysis, 191.Yohimbine, 16 1.Zinc, atomic weight of, 33.estimation of, 190

ISSN:0365-6217    

DOI:10.1039/AR9161300280

出版商:RSC    

年代:1916

数据来源: RSC

摘要:

284 ERRATA.ERRATA.ANNUAL REPOKTS. VOL. (1915).Page Line15G 20 The equation should read :6 6 C1 H C1\ /\ / I OH 0 OH ”OH C \OH/\OH HColourless product.I /\PH C HHAnthocyanin pigment frommyricetin.“ C.CH;C,H,( ObIe),. ”YeadMeO/\/\NANNUAL REPORTS. VOL. xrn (1916).Page 172, footnote (39) for “ Ll. Davi, ” yeutl “ W. ,4. Davis.’’,, 233, ,, (47) ,, ‘‘ J. 1%. Greaves ” read ‘‘ J. E. (;IY%~TTS.’’ ,, 247, line 1 G ” for ‘( 0.246” y e c d *‘ 0.26.”,, ,, ,, 5” ,, “ 0.238 ” ,, ‘ 6 0.24.”2 , 248, ,, 3 ,,“ 0.238 ” ,) ‘4 O.”.” 9 9 9 9 > ¶ 4 9 5( * 0.2446 ” ,, (‘ 0.26.’’* From below

ISSN:0365-6217    

DOI:10.1039/AR9161300284

出版商:RSC    

年代:1916

数据来源: RSC