CHEMICAL NEWS April 19, 1912 With such a method of attachment, when one arm is vibrated in a vertical direction, all the others move in unison. If one is moved in an upwards direction, then the two adjacent rods will move downwards, whilst that in the opposite position will move upwards (Fig. 4). 189 The difficulty with which cyclic compounds racemise may be illustrated by tying two opposite arms together, and re-adjusting the lengths of the strings which control the movement of the arms. It is difficult to bring the other two arms into the plane containing the asymmetric Differently coloured balls are attached to the arms, and carbon atom, since one of the balls must occupy a position a model of the asymmetric carbon atom is obtained. If within the ring. For a new arrangement to be formed the balls are placed in the order shown in the diagram this ball must pass through the ring, and this is presumably (Fig. 4), then the arrangement is that of an optically impossible if the group contains several atoms; thus the active compound, and if they occupy the vertices of a racemisation of a cyclic compound is practically impostetrahedron, the model now illustrates the van 't Hoff con- sible, unless it can take place through the formation of ception of the asymmetric molecule. an intermediate compound. The views of Werner may be illustrated by moving one of the balls up and down, thus causing the other balls to vibrate, This is taken to be the normal condition of the molecule. The application of some external strain, for Whilst the model does not explain the Walden inversion, it is useful in dealing with this process. In the model, only the simplest case can be represented, namely, that in which the amplitude of the vibration of all groups example, application of heat, causes an increase in the amplitude of the vibration of the groups until all lie in one plane for a small interval of time. This arrangement of the balls is not asymmetric, since a plane of symmetry can be drawn through the carbon atom. Once the balls are in this position there is an equal chance of their returning into their old arrangement or of passing into a new arrange. ment, which is the mirror image of the first; thus a ball, moving downwards, may continue its movement through the plane of racemisation, and vibrate about a new mean position. This change is similar to that which occurs in the so-called Walden inversion of optically active compounds. Since the new form so produced is transformed with the same facility into the old form, the model gives a method of illustrating Werner's view of racemisation. is the same, but this will not interfere with its use for purposes of demonstration. Another application of the model is to illustrate a possible explanation of the effect of the solvent on the of the rotation is thus represented by moving the arms rotation of an optically active compound. The diminution motion to them in this new mean position. The model nearer to the plane of racemisation, and imparting a may be used similarly to demonstrate in accordance with the above theory the changes in the rotation of an active compound with temperature. The re-adjustment of the strings connecting the arms will serve as a means of altering the valency directions, and so illustrate what may take place when one group is sub stituted by another. This will alter the direction of the plane of racemisation. (We are indebted to the Chemical Society for permission to reproduce the accompanying illustrations). *62. "Organic Derivatives of Arsenic and Antimony." (Preliminary Note). By GILBERT T. MORGAN and FRANCES M. G. MICKLETHWAIT. The Grignard reaction carried out with benzyl chloride and antimony trichloride leads to the production of tribenzylstibine dichloride (m. p. 105-108°), lustrous crystals, from alcohol; this is hydrolysed to tribenzylstibine oxide, which decomposes indefinitely at 240°. Other products of this condensation are being examined. Comparative experiments with arsenic trichloride and benzyl chloride have shown that the Grignard condensation leads to derivatives of quinquivalent arsenic, from which tribenzylarsine oxide has been obtained. Antimony trichloride has also been condensed under Grignard conditions with m-bromotoluene, p-dibromobenzene, p-iodonitrobenzene, 4-iodo-1 : 3-dinitrobenzene, and the three isomeric iodoanilines. These experiments are being continued with the object of preparing asymmetric organic derivatives of arsenic and antimony suitable for resolution into their optically active components. 63. "isoErucic Acid." By ALEXANDER MACBETH and ALFRED WALTER STEWART. KILLEN 64. "p-Hydroxystilbene and its Derivatives." Ву JOнн THEODORE HEWITT, WILLIAM LEWCOCK, and FRANK GEORGE POPE. The analogy existing between derivatives of azobenzene, benzylideneaniline, and stilbene has been extended by the preparation of 4-hydroxystilbene and its 4'-nitro-derivative, and comparison of their absorption spectra in neutral and alkaline solution with those of stilbene and 4-nitrostilbene. The corresponding methoxy-compounds have also been examined. 4-Hydroxystilbene is colourless, but shows a strong absorption band in the ultra-violet; a band which is partly in the visible region of the spectrum is shown by the 4'-nitro-derivative. Addition of alkali to the alcoholic solutions of both compounds causes the band to shift is considerably greater in the case of nitrohydroxystilbene towards the red end of the spectrum, but the displacement than in that of hydroxystilbene itself. 65. "The Chemistry of the Glutaconic Acids. Part V. The Esters of Substituted Glutaconic Acids." By NORMAN BLAND and JOCELYN FIELD Thorpe. It was shown that the esters of substituted glutaconic acids having two negative groups attached to the same carbon atom yield enolic sodium derivatives which dissociate in water to an extent depending on the tendency for the hydrogen atom to pass back into the three-carbon system. The esters formed in this manner are the normal esters of type (I.). The undissociated sodium compounds then yield the labile esters (type II.) when decomposed by carbon dioxide : The esters of type II. may exist in either their ketonic or enolic forms :— R.C.CO2Et CH CO,EACH-COJEt Ketonic form. on the nature of R. R.C.CO2Et || OH CO2E t.C:COEt Enolic form. In the course of some investigations of absorption spectra, the authors had occasion to purify brassidic acid, two specimens of which were obtained from Schuchardt. This acid was re-crystallised about fifteen times from alcohol, and had a sharp melting point (60°), which remained unchanged by the repeated re-crystallisations. On boiling it finally with animal charcoal, the meltingpoint was found to fall to 57-57'2°, and remained constant at this after further re-crystallisations. The absorp; and it was shown that the stability of these forms depends tion spectrum of the substance was examined, and found to show slightly less general absorption than either erucic or brassidic acid. Similar results are obtained if crude brassidic is used instead of the purified sample. An examination of the substance shows that it is unsaturated, as it is easily brominated in aqueous suspension, yielding a bromine derivative (m. p. 46-46.6°). This substance appears to be the dibromide of isoerucic acid (m. p. 42-43°), for the dibromide of brassidic acid melts at 54°. iso Erucic acid itself is stated to melt at 54-56°, so that it appears probable that the substance obtained as described above is isoerucic acid in a purer form than has hitherto been produced. The suggestion has been put forward in the case of oleic acid that the existence of the third isomeride is to be explained by assuming that in iso-oleic acid the double bond has been shifted into the a-position with regard to the carboxyl group. If the same idea is applied to isoerucic acid, we should expect to find its absorptive power to be greater than either that of brassidic or erucic acids, owing to the presence of a pair of conjugated double bonds in the molecule; but since it has actually less absorption than either of the other isomerides, this explanation of the isomerism cannot be applied to the present case. It is proposed to examine other cases of the kind with the view of determining whether the method is generally applicable to the preparation of these iso-forms, of existence in the free state; whereas when R = benzyl, Thus when R = methyl, the ketonic form is alone capable an equilibrium mixture of the two desmotropic individuals is produced. When a methyl group occupies the 8-position and a benzyl group the a-position, tautomerism ceases to be apparent, and the ketonic and enolic forms becomes structurally distinct. 66. O Viscosity of Compounds containing Tervalent Nitrogen. Part I. The Amines." By ALBERT GEORGE MUSSELL, FERDINAND BERNARD THOLE, and ALBERT ERNEST DUNstan. Forty-four amines have been examined in the pure state at 25° and 130°, also in amyl acetate and in hydrochloric acid solution. Viscosities have been expressed in terms of the quantity (n+106)/Mol. vol. The order of increasing viscosity is tertiary, secondary, primary. Substitution of an alkyl group for hydrogen lowers the molecular viscosity. Substitution of an acyl, phenyl, or benzyl group for hydrogen increases the molecular viscosity. Substances with a very high value of K have correspondingly high viscosities. There is a close parallelism between the association constant x and molecular viseosity. Meta-compounds have, in general, lower viscosities than ortho- or para CHEMICAL NEWS, Apparatus for obtaining an Average Sample of Gas. April 19, 1912 derivatives. Conjugation of unsaturated groupings increases viscosity, and acts in a similar way on K. "A Distillation Method for Separating Arsenic from Antimony and other Metals, and its Application to Toxicological Work." By STANLEY W. COLLINS. This method depends on the volatilisation of arsenic in the vapour of methyl alcohol and a stream of hydrogen chloride (compare Ber., 1895, xxviii., 1414), sulphuric acid being added in sufficient quantity to retain the more basic antimony. Compounds containing pentavalent arsenic are reduced by an excess of ferrous chloride before distillation. The method has been applied to the separation of arsenic from organic matters, including such small amounts as Uwould be present in toxicological work. As an example of this, estimations of arsenic have been made in the viscera 67. "Reciprocal Influence of Unsaturated Centres and its Effect on the General Absorptive Power of Compounds." BY ALEXANDER KILLEEN MACBETH, ALFRED WALTER STEWART, and ROBERT Wright. From an examination of the absorption spectra of various stereoisomerides, the authors find that if a molecule contains an unsaturated centre U, such as a carboxyl group, and a saturated centre S, such as an alkyl radicle, a variation in the relative spatial positions of S and produces very little change in the general absorptive power of the molecule. On the other hand, if the molecule contains two unsaturated centres, a change in their relative spatial positions produces a marked alteration in the absorption spectrum; thus erucic and brassidic acids have very similar absorption spectra, whereas the spectra of maleic and fumaric acid differ very considerably from each other. From this the authors deduce that in order to alter the general absorptive power of a molecule it is necessary to change the relative positions of at least two unsaturated centres within the molecule. The action of two centres of unsaturation is also shown by a comparison of the absorption spectra of ammonium thiocyanate and thiocarbamide. In the former there is only one unsaturated centre, namely, the CNS-group; whilst in the latter there are two, namely, the amido-radicles. The latter compound has a much greater absorptive power than the former. 68. “Experiments on a Yellow Colouring Matter from Ergot." By ALbert Freeborn. Another substance, It was shown that the three yellow colouring matters hitherto obtained from ergot (sclerocrystallin, ergochrysin, and secalonic acid) are in all probability identical, and have the composition C21H2209. crystallising in pale yellow needles, of the composition CISH1407, and melting at 338°, has now been obtained. It yields a tetra-acetyl derivative, C23H20O10, melting at 231°, and does not contain methoxyl. On fusion with potassium hydroxide, an unidentified phenolic acid is formed, probably containing at least three phenolic hydroxyl groups. 69. "Density of Acetic Acid. A Correction." By WILLIAM ROBERT BOUSFIELD and THOMAS MARTIN LOWRY. In the paper on acetic acid (Bousfield and Lowry, Trans., 1911, xcix., 1439) the difference in the density of acetic acid for 1° was given as 0.00123 instead of 0·00113. This does not affect the data given in the paper, but introduces errors of 0.0001 to 0.0003 in the reduction to 18° of the values given by other observers. SOCIETY OF PUBLIC ANALYSTS AND OTHER MESSRS. Maurice E. Balston, Howard A. Caulkin, and Certificates were read for the second time in favour of This paper deals with the composition of the milk of a small herd collected in September, 1911, and shows how the abnormal meteorological conditions affected the composition of the milk as regards non-fatty solid matters. from a case which had been treated with Salvarsan. "Estimation of Ferric Iron in presence of Organic Matter." By JOHN THEODORE HEWITT and GLADYS RUBY MANN. The authors find that ferric iron can be estimated in presence of organic matter by means of thiosulphate (Scherer, Kremer and Landolt, Oudemanns, F. Mohr, Braun, Joseph, Schatz). Three cc. of 1 per cent copper sulphate and I cc. of 1 per cent ammonium thiocyanate colour of the ferric thiocyanate is discharged, and the are added; N/10 sodium thiosulphate run in until the excess of thiosulphate determined by iodine and starch. Useful results were obtained in presence of alcohol, acetic, tartaric, and citric acids, phenol, and various carbohydrates. "The Relation of the Kirschner and Polenske Values in By E. RICHARDS BOLTON, H. DROOP RICHMOND, and Margarine containing Palm Kernel or Cocoanut Oils." CECIL REVIS. The Kirschner values for margarine containing cocoanut including 10 per cent, may be represented by straight line or palm kernel oil with or without butter-fat up to and curves for percentages between and including 5 and 75 per cent of these two fats. The presence of butter-fat displaces the curve laterally and practically parallel to the "no-butter" Kirschner curve, proportionately to the butter-fat present. is practically independent of the presence of butter-fat up The Polenske value for margarines containing these fats to and including 10 per cent, so that a mean value Polenske curve may be drawn which will be true for all mixtures, and acts as an "indicator value" determining the content of cocoanut or palm kernel oil in all cases. The curve is parabolic and the equations are given together with equations for estimating the amount of butter-fat from the Kirschner and Polenske values determined. A tentative relation between the Kirschner and Polenske values for butter-fat is given. "A convenient Apparatus for obtaining an Average Sample of Gas and for Regulating the Flow of a Gas into an Evacuated Vessel." By FRANK STURDY SINNATT. The apparatus consists of a gradually diminishing and adjustable mercury seal. The gas to be collected is forced to flow through the column of mercury. By this means samples of gases may be taken from any source at a constant rate, being collected in an evacuated vessel. The apparatus will find application in the obtainment of representative samples of gas over extended periods of time, as it requires practically no attention. The latter is a valuable feature, as other continuous sampling devices are not so simple nor so accurately adjustable. Comparative Hydrolysis of Saccharose by Different Acids in presence of Sucrose.-Gabriel Bertrand and M. and Mdme. Rosenblatt.—In the case of peroxydiastase, sucrose, and doubtless of many other soluble ferments, it appears that in presence of the specific colloidal substance of complementary activity, the activity of different acids does not depend only on the hydrogen ions which are due to their electrolytic dissociation, but also upon the nature of the radicles or anions to which the hydrogen is attached in the acid molecule.-Bull. Soc. Chim. de France, xi.-xii., No. 4. OBITUARY. DR, EDWARD DIVERS. It is with much regret that we announce the death at his London residence of Dr. Edward Divers, M.D., D.Sc., F.R.S. Dr. Divers was born in London in 1837; he received his early education at the City of London School, and subsequently studied at the Royal College of Chemistry and at Queen's College, Galway. After holding the post of Lecturer on Medical Jurisprudence at the Middlesex Hospital Medical School for three years, in 1873 he was appointed Professor of Chemistry at the Imperial College of Engineering in Japan, and became the Principal of the College in 1882. He was Vice President of the Chemical Society from 1900 to 1902, President of the Chemical Section at the British Association Meeting of 1902, VicePresident of the Institute of Chemistry in 1905, and President of the Society of Chemical Industry in the same year. He was elected a Fellow of the Royal Society in 1885, and received the Second Class of the Order of the Sacred Treasure and the Third Class of the Order of the Rising Sun in recognition of his services to scientific education in Japan. Dr. Divers' chemical work lay almost exclusively in the region of inorganic chemistry, and he was the author of many scientific papers, some of which were written in collaboration with his Japanese students. The compounds of nitrogen and oxygen were made the special subject of his study, and he published papers on the nitrites, the decomposition of hyponitrous acid and of its silver and mercury salts, and the existence and formation of salts of nitrous oxide. He investigated the constitution of nitrogen tetroxide, of the nitrososulphates of barium and lead, and of calcium sulphate, and also analysed many Japanese ores. In collaboration with Mr. T. Haga he studied the constitution and properties of nitrosulphonic acid, the preparation of hydroxylamine sulphate, and the reduction of nitrites to hydroxylamine, and with Mr. Kawakita published an account of the chemistry of the fulminates. Dr. Divers did some valuable work in the investigation of selenium and tellurium, and worked out a method of separating the two elements. CHEMICAL NOTICES FROM FOREIGN Comptes Rendus Hebdomadaires des Séances de l'Académie des Sciences. Vol. cliv., No. 7, February 12, 1912. Derivatives of Menthone.-Eyvind Boedtker.-By the action of alkyl magnesium bromides upon benzylidinementhone compounds of formula— may be obtained. The author has prepared thus menthomethylphenylmethane and menthoisoamylphenylmethane. Benzoylmenthone may be obtained by allowing benzoyl-chloride to act upon a solution of sodium menthone in toluene. The yield is poor. Lactonisation of a-Ketonic Ethers. Pyruvic Ether. H. Gault.-The constitution attributed by Genvresse to the ether resulting from the action of HCl on ethyl pyruvate is improbable, and the author believes it to be an enol ethoxy ether of ketovalerolactone carbonic ether. The action would then be similar to that of hydrochloric acid on pyruvic acid, which gives rise exclusively to ketovalerolactone carbonic acid. derivative analogous to the dibrom compound, and, with excess of chlorine, a product of formula C20H22Cl2O2 + 4CI, which is a tetrachloride of dichlorodehydrodicarvacroquinone. MISCELLANEOUS. Iron and Steel Institute.-The Annual Meeting of the Institute will be held at the Institution of Civil Engineers, Great George Street, Westminster, on Thursday and Friday, May 9th and 10th, 1912, commencing each day at 10.30 o'clock a.m. The following is the list of papers that are expected to be submitted for reading and discussion:"Notes on the Solubility of Cementite in Hardenite," by Dr. J. O. Arnold and L. Aitchison (Sheffield). "Chemical and Mechanical Relations of Iron, Vanadium, and Carbon," by Dr. J. O. Arnold (Sheffield) and Prof. A. A. Read (Cardiff). "Notes on a Bloom of Roman Iron from. Corstopitum (Corbridge)," by Sir Hugh Bell, Bart. (Middlesbrough). "Influence of Carbon on Corrosion," by C. Chappell (Sheffield). "Manufacture and Treatment of Steel for Guns," by General L. Cubillo (Valladolid, Spain). "Corrosion of Nickel, Chromium, and Nickel-chromium Steels," by Dr. J. N. Friend, J. Lloyd Bentley, and W. West (Darlington). "Mechanism of Corrosion," by Dr. J. N. Friend, W. West, and J. Lloyd Bentley (Darlington). "Sinhalese Iron and Steel of Ancient Origin," by Sir Robert A. Hadfield, F.R.S. (Sheffield). "Modern Rolling-mill Practice," by J. W. Hall (Birmingham). "Influence of Heat on Hardened Tool Steels," by E. G. Herbert (Manchester). "Improvements in Electric Steel Furnaces and their Application in the Manufacture of Steel," by Dr. H. Nathusius (Friedenshütte, Upper Silesia). "New Process for the Investigation of Fractured Surfaces of Steel," by F. Rogers (Sheffield). "Welding up of Blowholes and Cavities in Steel Ingots," by Dr. J. E. Stead, F.R.S. (Middlesbrough). "Note on some Remains of Early Iron Manufacture in Staffordshire," by Prof. T. Turner (Birmingham). The date and place of the Autumn Meeting will be announced at the Annual Meeting.-G. C. LLOYD, Secretary. MEETINGS FOR THE WEEK. TUESDAY, 23rd.-Royal Institution, 3. "Algernon Charles Swinburne -His Early Life and Work," by Edmund Gosse, Faraday Society, 8. General Discussion on "Magnetic Properties of Alloys." (See p. 180). WEDNESDAY, 24th.-Royal Society of Arts, 8. "Technical Education in Ireland," by G. Fletcher. THURSDAY, 25th.-Royal Institution, 3. "Synthetic Ammonia and Nitric Acid from the Atmosphere," by Prof. A. W. Crossley, F.R.S. Royal Society of Arts, 4.30. "The Central Provinces," by Sir John O. Miller, K.C.S.I. Royal Society. "Diffusion and Mobility of Ions in a Magnetic Field," by J. S. Townsend "Observed Variations in the Temperature Coefficients of a Precision Balance," by J. J. Manley. "The Torque produced by a Beam of Light in Oblique Refraction through a Glass Plate," by Dr. Guy Barlow. "Contributions to the Study of Flicker," by Dr. T. C. Porter. Mr. H. FRIDAY, 26th.-Royal Institution, y. "Sir William Herschel," by Sir George H. Darwin, K C.B., &c. Physical, 5. Adjourned Discussion on Donaldson's paper on "The Coefficients of Expansion of Fused Silica and Mercury." "Solution of Net-work Problems by Determinants," by R. Appleyard. Method of Measuring Small Inductances," by S. Butterworth. Action of Bromine and Chlorine on Dehydrodicarvacrol.-H. Cousin.-Bromine reacts with dehydrodicarvacrol to give a dibrominated derivative, C20H24Br202. SATURDAY, Chlorine yields two different compounds, the dichlor 27th.-Royal Institution, 3. "The Architecture of the Renaissance in France," by Reginald Blomfield. CHEMICAL NEWS, April 26, 1912 Red Water. 193 remain. It does not seem possible that the selection can THE CHEMICAL NEWS. be of an arbitrary nature, and one therefore looks for an VOL. CV., No. 2735. explanation. Since it must be assumed that in a gaseous mixture the different molecules are equidistant from one another, it is impossible to afford an explanation on the basis that some of the molecules are more advantageously situated for APPARATUS TO STUDY THE DIFFUSION OF combination than others. It is well known that when two gases combine together they do so in some definite proportion, and that if a mixture of the two gases is taken in that proportion and combination effected, the resulting product will consist only of the compound formed by their union. If, however, a mixture be taken such that there is an excess of one gas above the amount required for the formation of the particular compound, then, after action has taken place, the residue consists of the compound of the two gases together with an uncombined residue of that gas which was originally present in excess. The question therefore arises :-What is it that determines which molecules of the gas in excess enter into combination, and which are left unchanged at the end of the reaction. If, for example, a mixture of N molecules of hydrogen and N molecules of oxygen be taken and union effected, then the product consists of N molecules of water together with N/2 molecules of uncombined oxygen; the hydrogen molecules have thus in some way exerted a selective influence, and have combined with N/2 molecules of oxygen, thus leaving a further quantity of N/2 molecules of oxygen uncombined. What therefore is the factor that determines which of the molecules of oxygen shall enter into combination? There must have been in the original mixture some difference between the N/2 molecules which have gone to form water and the N/2 molecules which The kinetic theory of gases, however, at once suggests a possible explanation; for since the molecules of a gas are moving with different velocities, and since combination depends on the number of impacts made between different molecules in a given time, then it follows that the molecules of a gas which are moving with the greater velocity will be those which enter into combination, while the molecules which are moving with the less velocity are left. should possess a lower temperature than would be antiThis explanation at once suggests that the resulting residue cipated, since the temperature of a gas depends on the mean velocity of its molecules. It is possible that this phenomenon might be capable of experimental verification. Thus in one experiment two gases would be taken in combining proportions, and the temperature of the product taken after combination had taken place; in another experiment a mixture of the two gases would be taken such that there was an excess of one gas present above that required for the formation of the compound. Then if the temperature of the product is observed after interaction has taken place, and the value obtained corrected for dilution of the mixture, it would be expected that the latter temperature would be less than the former owing to the removal in one case of the molecules possessing the greater velocity. Again, the temperature might be observed after combination when the two gases are present in combining proportions, but diluted in one experiment with a certain volume of an inert gas, and diluted in another experiment with the same volume of one of the interacting gases. After due corrections have been applied the latter temperature should be less than the former value, owing to the fact that in the former experiment no selective removal of molecules can take place. It is, of course, most probable that the difference in temperatures recorded would be very small; however, the kinetic theory of gases demands, or at least suggests, that some such difference should be found. A SAMPLE of water sent by Dr. R. van Someren, Kystume SS., comp. by Kompola, Uganda, for analysis, has provided results which may be of some interest. In the letter accompanying the sample he says: "It comes from a crater lake out here, and I am very curious to know what gives it the red colour. The lake looks like blood at times. It seems to be saturated NaCl with some (?) lime." The sample received had a distinct rose-pink colour. The bottle containing it was about three-quarters full of liquid, and a small amount of crystalline solid had separated, probably owing to evaporation and lowering of temperature. The red colour was not separated from the water by filtration through ordinary filter-paper, but it was removed by means of a Berkefeld filter. On microscopic examination of the red deposit on the Berkefeld candle only disintegrated organic remains were seen. When mixtures of equal quantities of the water and nutrient agar were incubated at 25° and 30° C. there appeared on the sloped surface white growths of bacteria, which did not, however, |