ii. 20 ABSTRACTS OF CHEMICAL PAPERS. Mineralogical Chemistry. The Calculation. and Comparison of Mineral Analyses. C. E. VAN ORSTRAND and FRED. E. WRIGHT (J. Wushington Acad. A%. 1914 4 514-525).-A mathematical discussion and criticism of the papers by Schaller and Wells (ibid. 1913 3 97 416) on the calculation of mineral f ormulz and the interpretation of mineral analyses. I n all cases the method of least squares is involved. Several equations and formulze itre given f o r the computation of the probable errors and the adjustment of the data. Identity of Empressite with Muthmannite. WALDEMAR T. SCHALLER (J. Fashinylon Acad. A% 1914 4 497-499).-The two minerals empressite (Bradley A. 1914 ii 737) and muthmannite (Zambonini A. 1911 ii 734) are considered by the author t o be really the same substance.The physical characteristics are very similar the only difference being that whilst empressite is AgTe and contains no gold muthmannitei contains considerable quanti- ties of gold and is expressed by the formula (AgAu)Te. The quantity of gold is however always less molecularly than the silver; the author therefore expresses the opinion that empres- site is only pure muthmannite and could be more accurately named muthmannite whilst the original mineral should be named auric muthmannite. J. F. S. L. J. S. Analytical Chemistry. Symbolical Representation of Analytical Operations. 1,. GOWING-SCOPES (C'hem. News 1914 110 247-250).-8 list of some ninety symbols for representing analytical operations is given ; Dihydroxydistyryl Ketone or Lygosin a New Indicator.ARON FERENEZ (Ann. Chim anal. 1914 19 232 ; from Phnrw. Post. 1913 521).-When 2 mols. of salicylaldehyde and 1 mol. of acetone in alcoholic solution are treated with concentrated sodium hydr- oxide soslution condensation takes place with the formation of a salt which may be crystallised from alcohol; this salt on treat- ment with an acid yields o-dihydroxydistyryl ketone a substance giving an orange-red coloration with alkalis and a turbid opal- escence with acid solutions. The indicator may be used in the form of a 1% alcoholic solution; it is stated to be as sensitive as phenolphthalein and may be employed in the titration of ammonia. w . P. s. their use is indicated by a number of examplw. 'cv. P. S.ANALYTICAL CHEMISTRY. ii. 21 Methyl-red a8 an Indicator.R . T. TROMSON (Analyst 1914 39 518-519).-This indicator is especially useful in the titration of ammonia and weak bases such as quinine (compare A 1909 ii 90); its colour change is sharper than that of methyl-orange and its sensitiveness is but little affected by the presence of neutral salts such as sodium chloride or sulphate. It must not be assumed that methyl-red can be substituted for methyl-orange or that it will act similarly t o the latter; methyl-red is affected by dissolved carbon dioxidel and when used as the indicator in the titration of carbonates the solution must be boiled after each addition of acid in order to expel the liberated carbon dioxide. Methyl-red in this respect most closely resembles litmus (compare also T.w. P. s. Qualitative Test for Water by the Use of the Acetylene- Cuprous Chloride Reaction. E. R. WEAVER (J. Amer. Chum. Xoc. 1914 36 2462-2468).-The presence of water in a substance can be readily detected by bringing it into contact with calcium carbide and a solvent for acetylene and then decanting or distil- ling the resulting solution into an ammoniacal solution of cuprous chloride. Most of the usual organic solvents if carefully dried are suitable for this purpose. The method is applicable t o a great variety of substances and especially to the volatile organic com- pounds. The only compounds which interfere with the test are the stronger acids and substancw such as hydrogen sulphide which precipitat.e cuprous salts from solution. The test is sensitive to less than 0.1 mg.of water. 1910 97 2477). E. G. Rapid Methods of Estimating Water in Crude Petroleum Oil-Fuel and Similar Substances. HERBERT S. SHREWSBURY (Analyst 1914 39 529-831).-A simple method is described for estimating the water by distillation. One hundred C.C. of the oil and a few pieces of pumice-stone are placed in a distillation flask which is so supported that its sideltube passes vertically down- wards into a graduated cylinder surrounded by cold water. After the neck and surface of the flask (uncovered by the oil) have been heated cautiously with a flame the oil itself is heated until all the water and a few C.C. of oil have distilled over into the cylinder and the volume of the water is read. Another method depends on the extraction of the water frosm oil-fuel by glacial acetic acid and on the alteration of the turbidity temperature of the acid with a standard oil caused by the presence of the water.The turbidity temperature of the acetic acid with the oil (equal volumes of arachis and cocoanut oils) is ascertained; 10 C.C. of the stan- dardised acid and 10 C.C. of the d-fuel are then shaken together the acid layer is separat'ed filtered and 2 C.C. of the1 filtrate are lieatled with 2 C.C. of the standard oil until the mixture is clear and its turbidity temperature determined. Standards must be prepared by adding known quantities of water to the acetic acid. A correction is necessary for the effect on the turbidity temper ature of substances other than water dissolved from the oil-fuel;ii. 22 ABSTRACTS OF CHEMICAL PAPERS.bhis is obtained by drying a portion of the oil-fuel by heating it in an open basin and then shaking 2 C.C. of it with glacial acetic acid as described. When the oil-fuel contains more than 2.5% of water the acetic acid extract must be diluted with acetic acid exactly the same dilution being made in the case of the extract giving the figure for the correction. Both methods can probably be used for the estimation of water in substances such as butter lard and other fats and oils. w. P. s. Examination of Sulphur Chloride. F. FRANK and E. NARCRWALD (Gummi-Zeit. 1914 28 1580-1581 ; from J. Xoc. Chom. Ind. 1914 33 749).-Whilst pure sulphur chloride has D15 1.684 samples having D15 1.680 +o 1.690 may be considered as satisfactory. When 105 C.C.are distilled using a standard flask and condenser (of which minute details are given) 95 C.C. should distil within 4O the first 5 C.C. being rejected. The sample should contain at least 98.8% of S,Cl calculated on the chlorine content estimated by the following method From 10 to 15 C.C. of nitric acid (D 1.5) and a quantity of silver nitrate sufficient for the reaction are1 introduced into a small pressure flask having a ground-in stopper secured by a lever. A small weighing bottle contlaining from 0.2 to 0.5 gram of the sample is then dropped into the flask this is shaken until all the sulphur chloride has dis- appeared aad is then heated for thirty minutes on a water-bath cooleld and opened. The precipitate is collected dissolved in ammonia re-precipitated again collected and weighed.The method yields trustworthy results higher than those obtained by the older process. w. P. s. Modifled Kjeldahl Flask for Estimating Soil Nitrogen. H A. NOYES (J. Amar. Chern. Soc. 1914 36 2541-2542).-As the amount of nitrogen in ordinary soil is small a large sample must be employed for its estimation and this usually necessitates the transfereace of the solution from the digestion flask to another flask for distillation t o avoid bumping. I n order t o obviate this transference a modified Kjeldahl flask may be used which has a capacity of about 700 C.C. and differs from the ordinary form in being egg-shaped. The pointed bottom of the flask prevents the solid matter from settling during the heating and frothing does n o t occur as the bubbles are broken against the sides of the1 flask.The flask has proved very satisfactory in practice. E. G. Preparation of Reduced Strychnine Reagent for the Colorimetric EBtimation of Nitrates in Water. G. DENIG~S (Ann. Chime anal. 1914 19 221-222 ; from Bull. Soc. pharm. Bordmux 1914. Compare A. 1911 ii 655 673).-The zinc amalgam used is obtained by immersing 50 grams of zinc foiI cut into small strips in a mixture of 50 C.C. of water and 20 C.C. of hydrochloric acid (D 1-18) and then adding 10 C.C. of 4% mercuric chloride solutio'n; after two minutes' contact the strips are washed vith water placed on a sheet of filter paper until dry and thenBKALY'I'ICAL CHEMISTRY. ii. 23 stored in a closed bottle. The reagent itself is prepared immediately before use as described (Zoc.c i t . ) . The red coloration produced by the action of nitrates on the reagent in the presence of sulphuric acid attains its full intensity only after the reacting mixture has been heated on a boiling water-bath for five minutes. Estimation of Phosphorus in Phosphor-Bronze. E. W. HAGMAIER (Net. aid Chem. Eng. 1914 12 524-525; from J. SOC. Chem. Ind. 1914 33 923-924).-0ne gram of the bronze is dissolved in a mixture of 10 C.C. of hydrochloric acid and 5 C.C. of nitric acid the solution is diluted with 150 C.C. of hot water 10 C.C. of a solution containing 1 gram of cerium chloride in 25 C.C. of hydrochloric acid and 200 C.C. of water are added and ammonia is run in slowly from a burette until a greenish-blue coloration appears. Four C.C. of acetic acid are now added and the mixture is boiled for ten minutes.After settling the clear liquid is siphoned off the precipitate washed by decantation with 100 C.C. of hot water (added in about six successive portions) and dissolved by heating with 15 C.C. of nitric acid and 3 C.C. of hydrochloric acid. When cold the solution is treated with 5 C.C. of ammonia (1 l) and the phosphoric acid precipitated in the usual way with Estimation of Boric Acid and its Compounds in Extract of Tomatoes and Other Preserved Foods. Nrco~lis CAMUS (And. SOC. Quirn. Argentina 1914 2 123-130).-After ashing the residue from which all the carbon need not be burnt off is moistened with strong hydrochloric acid and the boric acid ex- t,racted with acetone. After evaporation of the solvent the resi- due is dissolved in about 3 C.C.of very dilute hydrochloric acid and the boric acid recognised by means of turmeric paper. By standardising the conditiions size of paper strength of turmeric etc. it is possible to obtain approximately quantitative values. W. P. S. ammonium molybdate. w. P. s. G. D. L. Estimation of Cadmium in Zinc. W. COOPER (Chem. News 1914 110 250-251).-Five grams of the sample1 are dissolved in a cold mixture of 150 C.C. of water and 7 C.C. of sulphuric acid and the solution is decanted into a flask any undissolved residue being washed with a small quantity of water. Twenty C.C. of dilute sulphuric acid ar0 then added to the solution the mixture is diluted t o 250 c.c. treated with hydrogen sulphide for twenty minutes and after two hours the precipitated cadmium and zinc sulphides are collected on a filter.Ths sulphides are now dissolved in hydrochloric acid containing bromine 5 C.C. of sulphuric acid are added and the mixture is heated until fumes of sulphuric acid are given off. This residual solution is diluted t o 50 c.c. any lead sulphate is separated and the precipitation with hydrogen sulphide and treatment with sulphuric acid a,re repeated. The solution is once more treated with hydrogen sulphide the cadmium sulphide is collected on a weighed filter washed dried a t looo and weighed. w. P. s.ii. 24 ABSTRACTS OF CHEMICAL PAPERS. Iodate Method for the Estimation of Copper. W. W. BROSTROM (Eng. and N i n . J. 1914 98 215-216; from J. Scc. Cham. Ind. 1914 33 891).-The copper is reduced with sodium sulphite solution in the presence of hydrochlo,ric acid and pre- cipitated with potassium thiocyanate.The precipitate is washed decomposed with hydrochloric acid in the presence of a small quantity of carbon tetrachloride and the solution titrated with potassium iodate solution until the red colour produced a t first is discharged. Results are given showing that the method is trust- Estimation of Cuprous and Cupric Sulphides in Mixtures of One Another. EWEN POSNJAK (J. Amer. Chern. ~ u c . 1914 36 2475-2479).-During an investigation of copper sulphide minerals it was desired to estimate cuprous and cupric sulphides in presence of one another. A method has therefore been devised which depends on the difference in the behatriour of the sulphides towards silver nitrate as shown by the equations Cu,S + 4AgNO,= Ag,S + 2Ag -I- 2Cu(NO,) and CuS + 2AgNO,=Ag,S + Cu(N03),.I n order to carry out an estimation the substance must be fine enough to pass through a sieve of 80 meshes per cm.. About 0-4-0.5 gram of the material is treated with 50 C.C. of 5% silver nitrate solution which is added gradually with vigorous stirring. To complete the reaction the mixture is heated on the steam-bath for three hours and stirred vigorously a t short intervals. The precipitate is collected and washed and then treated two or three times with 6% ferric nitrate solution a t 70° to extract the silver. The extract is filtered and the residue washed with hot water. Nitric acid is added t o the filtrate until it becomes colourless and the silver is then precipitated and weighed as the chloride.The amount of cuprous sulphide is calcu- lated from the silver found whilst the cupric sulphide is calcu- lated from t'he difference bet'ween the silver in the silver sulphide and that precipitated in the metallic form. The method has been found t o be accurate within 1.5% for the analysis of mixtures of the sulphides in any proportions. worthy. w. P. s. E. G . Electrolytic Estimation of Nickel. W. JUDSON NARSH (J. Plip~icd Chem. 1914 18 705-716).-ft is shown that when nickel is estimated electrolytically by deposition from solutions containing an excess of ammonium oxalate the weight of the d e posit is always too high owing t o the presence of carbonaceous material. The present paper deals with experiments made with the object of improving the process.The nickel deposit is shown to contain nickel carbide which is supposed to be formed by the action of acetylene on the freshly deposited metal. The acetylene is produced by the cathodic reduction of carbon monoxide and carbon dioxide formed at the anode. Attempts were made t o reduce the amount of carbon dioxide formed by adding reducing agents to the solution such as hydrazine sulphate hydroxylamine sulphite sodium nitrite and sodium sulphite. Of these substances,ANALYTICAL CHEMISTRY. ii. 25 the first only produced a noticeable effect inasmuch as it caused the deposit to be much darker probably due to the formation of nickel sulphide. This conclusion is rendered likely because when a nickel solution was electrolysed in the presence of ammonium carbonate and sodium sulphite a flocculent precipitate of nickel sulphide was produced and hydrogen sulphide was evolved.Having failed tlo improve the oxalate method of estimation the author has investigated a number of other methods and comes t o the conclusion that the electrolysis of the double ammonium sul- phate solutioas of nickel gives the best results. A solution of nickel sulphzcte containing about 0.2 gram of nickel in 200 C.C. is mixed with 15-20 C.C. of ammonium hydroxide solution (D 0.96) and 5 grams of ammonium sulphate. The electrolysis is effected between an anode of platinum gauze 6 cm. x4.5 em. and a cathode of copper gauze made in the form of a cylinder 5 cm. high and 3 cm. diameter. The cathode is rotat'ed a t 550-6563 revolutions per minute.The electrolysis is carried out a t ordinary temperature with a current of 0.8 ampere and an E.M.F. of 3.5 volts and requires about twenty-five minutes. It is essential that nitrates and chlorides. should be absent. J. F. S. Modificatione of the Reduction Test for Tungstic Acid. G. TOROSSIAN (Amer. S. Sci. 1914 [iv] 38 537-638).-Wheri powdered tungstic acid is placed on aluminium foil and moistened with a drop of dilute hydrochloric acid a blue coloration is obtained; the reaction is also given by thungstic acid after fusion with sodium carbonate and the presence of ferric oxide copper oxide lead oxide sulphur calcium fluoride or sodium nitrate does not interfere with the test. w. P. s. Assay of Crude Platinum. M.SCHWITTER (Eng. and Nin. J. 1914 97 1249-1250; from J. S'oc. Chem. In& 1914 33 751).- Twenty grams of the nuggets are digested with 200 C.C. of aqua regia (lHNO 3HC1) on a water-bath f o r one hour and then boiled f o r two hours. The solution is decanted and again boiled to dissolve any finely divided metal in suspension. The aqua regia treatment is repeated on the residue the mixture is diluted and allowed to settle. After about eighteen hours the clear solu- tion is decanted and the last portion filtered into a litre flask the filter being washed until colourless. Osmn'ri&um.-The residue from the acid treatment is scorified with 10-15 grams of test lead and enough litharge to fill the scorifier. The button is cleaned with as little hammering as possible and dissolved in dilute nitric acid (1 3) ; the solution contlaining the lead is filt'ered the filter washed free from lead and dried.The osmiridium is separated as completely as possible from the paper the latter is burned and the ash added to the metal which is weighed. Platinum.- Twenty-five C.C. of the filtrate are evaporat'ed to dryness the resi- due heated for one hour with occasional stirring wilh 5 C.C. of hydrochloric acid (1 9) 20 C.C. of water and 8 grams of ammonium chloride and then set aside for eighteen hours; theii. 26 ABSTRACTS OF CHEMICAL PAPERS. platinichloride is collected washed with 20% aniinonium chloride solution and ignited in an annealing cup. The metal contains most of the iridium; the quantity of the latter if not more than 5% can be judged with sufficient accuracy from the colour of the platinichloride 0.2% of iridium sufficing to impart a decided red tint.Gold.-The filtrate from the platinichloride is treated with 5 grams of ferrous sulphate the gold collected and ignited. Palladium.-The filtrate from the gold is treated with one-fifth of its volume of hydrochloric acid and 10% potassium iodide solu- tion is added five drops at a time until no further quantity of black precipitate (PdI,) forms. The mixture is boiled cooled the precipitate collected washed with dilute hydrochloric acid (1 4) until free from iron ignited in a porcelain crucible then heated with a few drops of formic acid to reduce any oxide dried and weighed. Rhodium.-The filtrate from the palladium is reduced while hot with zinc and the metallic precipitate is digesteld for ten minutes on a water-bath with 25 C.C.of dilute nitric acid (1 3) ; the residue is ignited reduced in hydrogen and weighed as rhodium; it contains the last traces of iridium. w. 9. s. Detection and Estimation of Combustible Gaaes in Mineral Waters. ENRIQUE HAUSER (Anal. Pis. Quim. 1914 12 155-1 63 256-259. Compare A. 1914 ii 285).-A minute description of the eudiometer devised by the author for the estimation of com- bustible gases in water and of the mode of operating it. Th& gas is ignited electrically in the eudiometer and the products are analysed by means of ordinary absorption pipettes. The solubility of methane varies according to the nature of the saline solution. A. J. W. Rapid Method for the Detection and Estimation of Solid Hydrocarbons Mixed with Fatty Acids.G. VERONA-RINATI (Ann. Chim. Appticnta 1914 2 20L-202).-This method is carried out as follows. By means of a buretbe 7 C.C. of dry a-dichloro- hydrin b. p. 174O are measured into a glass tube having a diameter of about 10 mm. and a capacity of about 15 c.c. and graduated to read easily t o 0.1 C.C. To this are added 1 . 2 C.C. of water and 1.43 grams of the substance t o be tested. The tube is closed tightly by means of a cork and immersed in a beaker of water maintained a t 65* being removed and well shaken from time t o time. When the liquefied mass is no longer diffused uniformly throughout the whole of the liquid the1 tube is left at rest for ten minutes and the volume of the1 undissolved part collected a t the top is read; this reading is repeated after further heating and shaking.Each 0.1 C.C. of this volume corresponds with 5% of fatty matter of mineral origin in the original substance. I n order t o ascertain whether such mineral fat consists of cerasin o r paraffin or a mixture of these about 2-3 grams of the original material are treated in a test-tube with the aqueous a-dichlorohydrin until the fatty acids are completely dissolved. The tube is then allowed t o cool until the undissolved portionANALYTICAL CHEMISTRY. ii. 27 becomes consolidated a t the top whilst the fatty acids remain in solution in the liquid underne'ath. The solid mass is removed and pressed between filter papers tol free it from the solvent; 0.2 gram of it is heated in a test-tube over a naked flame with 10 C.C.of anhydrous a-dichlorohydrin and the temperature is determined a t which the dissolved hydrocarbon begins to' separate again and render the liquid turbid. When this hydrocarbon consists solely of paraffin this temperature never exceeds 92O (88-92O) whilst with cerasin it is never below 1 1 2 O (l12-l15°). With a mixture of paraffin and cerasin the temperature of separation lies between 92O and 112O and its value indicates which of the two pre- dominates in the mixture. T. H. P. Estimation of Sucrose and Lactose in Condensed Milk by means of Acid Mercuric Nitrate. CECIL REVIS and GEORGE A. PAYNE (Analyst 1914 39 476-479).-The acid mercuric nitmte solution employed is prepared by dissolving mercuric oxide in twice its weight of nitric acid I3 1-42 and diluting the solution t o five times its volume with water A quantity of 65.085 grams of con- densed milk is dissolved in about 220 C.C.of water heated in a boiling water-bath for about five minutes (or kept cold for sixteen hours) and then diluted to 250 C.C. a t 1 5 O . One hundred C.C. of this solution are transferred t o a 100 C.C. cylinder 10 C.C. of the mercuric nitrate solution are added the mixture is shaken for thirty seconds filtered after five minutes and the filtrate is polarised a t 20O. Twenty-five C.C. of the filtrate are then placed in a 50 C.C. stoppered flask the flask and its contents are weighed heated in a boiling water-bath for eight minutes cooled the weight is adjusted and the solution polarised.The readings observed are corrected for the increase in volume due to'the added reagent less the volume of the f a t and protein precipitated; this correc- tion (C) will be and the readings are mukiplied by (lOO+C). Then sucrose %= (D-T)100/(E,-t/2) and lactose1 %=((u-S) x RL1x 100/26-034 where B and Z are the corrected direct and invert readings R and RL the inversion and lactose factors and t the temperature of the invert reading. The factors R and R were' found to be 141.71 and 0.3086 respectively; these figures are the results of actual experiment with sugar solutions of the concentration met with in the case of condensed milks thel authors having ascertained that acid mercuric nitrate has an inverting action on sucrose and that it slightly raises the1 rotlation of lactose.When the milk is clarified with phosphotungstic acid the factars are found to be 142.55 and 0.3165 figures which agree with those recorded by other observers Flour. I. Acidity of Flour. 11. Natural and Artificial Bleaching of Flour. 111. Sulphates and Lime in Flour. R. T. THOMSON (Analyst 191 4 39 519-527).-The aqueous extract of flour is acid to phenolphthalein neutral t o methyl-red C z= 10 - [(P x 1.11) + (P x 0*82)/100 x 26.0341 c.c. (compare A. 1907 ii 586). w. P. s.ii. 28 ABSTRACTS OF CHEMICAL PAPERS. and alkaline to methyl-orange. The acidity indicated by phenol- phthalein is not due to the presencc of lactic acid since Lhis acid is also indicated by methyl-red and i t is doubtful whether lactic acid is the acid produced when flour becomes “sour.” There is no support for the usual statement that ordinary flour is acid in reaction.The author gives the results of experiments showing that artificial bleaching of flour by nitrogen peroxide does not have the same effect as natural bleaching that is by exposure to the atmosphere. Although flour absorbs nitrogen oxides from the air the bleaching effect in this case is due to other causes probably oxidation assisted by moisture and light. For the estimation of sulphates in flour it is recommended that 20 grams of the flour should be mixed with 250 C.C. of water containing 15 C.C. of hydro- chloric acid D 1.16 and then heated until the1 starch is liquefied. The mixture is cooled filtered and the sulphates are precipitated in the filtrate by the addition of barium chloride.First-grade flour was found to contain 0.011% of SO and 0.015% of CaO whilst common-thirds ’’ yielded 0.061% of SO and 0.102% of CaO (com- Utz’s Modiflcation of Halphen’rs Reagent for Cottonseed Oil. ENRICO GASTALDT (Ann. Chim. Appliczta 1914 2 203-207).- At no temperature does Utz’s reaction (compare A. 1914 ii 591) give the same coloration with cottonseed oil as Halphen’s reagent and as i t is exhibited a t 250-260° a t which temperature Halphen’s test fails it cannot be regarded as a modification of the latter. A t a temperature exceeding 160° which is usually sur- passed in Utz’s reaction oleic palmitic or stearic acid gives a colora- tion similar t o that obtained with cottonseed oil. This reaction cannot therefore be regarded as characteristic for cottonseed oil.The author’s modification of this test (compare A. 1912 ii 1108) may be carried out very rapidly and reveals the presence of small proportions of cottonseed oil in the following manner five to1 six drops of a 1% solution of sulphur in carbon disulpliide and three to four drops of pyridine are added to 5 C.C. of the oil or fat and the liquid heated directly over a flame for four t o five minutes care being taken that the temperature does not exceed 140O. T. H. P. pare A. 1914 ii 816). w. P. s. The Freezing Point of Milk. J. BROWNLIE HENDERSON and L. A. MESTON (Chem. News 1914. 110 259-261 275 283-284 ; from Proc. Roy. Soc. QueenslacncF 24 165).-The freezing point of fresh mixed milk from a herd of cows varies from -0*55O t o -OV56O the average being -0.555O.The constancy of the freez- ing point affords a means of detecting the presence of added water in milk since a rise of 0*005O indicates the presence of about 1% of water. The freezing point is not however a measure of the proportion of total solids or of non-fatty solids in the milk. As the milk becomes acid the freezing point is lowered but the authors are not yet in a position t o state whether any definite relation exists between the depression of the freezing point andANALYTICAL CHEMISTRY. ii. 29 the increase of acidity. Tables are given showing C.C. of added water per litre to milk corresponding with temperatures ranging from -0.55" t o -0.351" for differences of O*O0lo and ths method of determining the freezing point is described in detail.JV. P. S . [Determination of the] Freezing Point of Milk. G. A. STUTTERHEIM (Phacrm. Weekblad 1914 €5 1 131 1-1 313).-A descrip- tion oE a modified form of Beckmann apparatus for the determina- tion of the freezing point of milk (co,miire A. 1914 ii 169 304). A. J. W. Estimation of the Cineole (Eucalyptol) Content of Euca- lyptus Oils. HENRY G. A. HARDING (Analyst 1914 39 475-476). -The resorcinol method (compare A. 1908 ii 233) for the esti- mation of cineole is unt'rustworthy when applied directly t o euca- lyptus oil but better results are obtained by using the fraction of the oil b. p. 170-190°. for the test. Even then the results tend to be too high especially in the case of oils containing more than 50% of cineole. Turpentine has been suggested as a diluent but since it1 contains substances which are absorbed by resorcinol only that portion should be used which distils between 156O and 160O.One hundred C.C. of the eucalyptus oil under examination are dis- tilled the fraction distilling between 170° and 190" is collected separately and diluted t o 100 C.C. with the turpentine fraction mentioned; a preliminary test will show whether further dilution is necessary to reduce the cineole content t o 50%. Ten C.C. of the solution are then shaken in a flask having a graduated neck with a warm 55% resorcinol solution for five minutes a further quantity of the resorcinol solution is then added so as to bring the oil into the neck of the flask and after cooling the volume of the oil is read. w. P. s. Estimation of Camphor and of Certain Essential Oils when in Solution in Alcohol.W. B. D. PENNIMAN and W. W. RANDALL ( J . I n d . Eng. Chem. 1914 6 926-928).-The method proposed depends on the fact that camphor and the essential oils of pepper- mint lemon orange anise and nut'meg are liberated from their alcoholic solution when this is mixed with from four to ten times its volume of concentrated calcium chloride solution and that the separated camphor or oil is soluble in light petroleum b. p. 40-60°. Within certain wide limits the increase in volume of the light petroleum is equal t o the volume of the camphor or essential oil. A measured quantity of the alcoholic solution under examination is mixed in a flask provided with a graduated neck with calcium chloride solution a definite volume of light petroleum is added (about equal t o the volume of the oil to be dissolved) more of the calcium chloride solution is added t o bring the mixtare into the neck of the flask the whole is submitted t o centrifugal action and the increase in volume of the light petroleum noted.W. P. S.ii. 30 ABSTRACTS OF CHEMICAL PAPERS. Detection of Emodin-bearing Drugs in the Presence of Phenolphthalein. L. E. WARREN (Amer. J. Phovm. 1914 86 444-449).-The red coloratioii obtained when emodin is treated witah an alkali cannot be used for the detection of this substance in mixtures also containing phenolphthalein but if the latter is converted into tetraiodophenolphthalein the emodin may be ex- tracted with chloroform and identified by the alkali test.The substance is extractled with acetone or chloroform the extract is evaporated the residue dissolved in dilute sodium hydroxide solu- tion and a slight excess of iodine solution is added. The mixture is then rendered acid with hydrochloric acid the insoluble tetra- iodophenolphthalein is separated by filtration and the filtrate after the addition of sodium sulphite is shaken with chloroform; this chloroform extract is evaporated and the residuel tested with alkali hydroxide solution. I n the case of preparations containing phenolphthalein a faint red coloration may sometimes bet obtained but' this disappears on the addition of an excess of the alkali whilst the coloration given by eniodin is permanent. The emodin colora- tion is of various shades of red according t o the source of the emodin.Except in the case of aloes the method is trustworthy. A method described recently by Bailey (A. 1914 ii 501) was also Estimation of Prussian-blue in Tea. G. W. KNIGHT (J. Z7,d. Eng. Chern. 1914 6 909-910).-The following method is recom- mended for the estimation of very small amounts (say 1 part per 200,000) of added Prussian-blue in tea. One hundred grams of the powdered sample are moistened with 85%. phosphoric acid and distilled until phosphoric acid begins tor pass over the distillate beiBg collected in a receiver containing dilute sodium hydroxide solution. The distillate if acid is neutralised 3 C.C. of 10% sodium hydroxide are added in excess then a small crystal of ferrous sulphate and a few drops of ferric chloride solution and the mixture is boiled for one minute.The solution is then acidified with hydrochloric acid the precipitate collected washed with alcohol and dissolved in a small quantity of sodium hydroxide solution. After this solution has been acidified with acetic acid a few drops of hydrochloric acid and ferric chloride solution are added followed by more hydrochloric acid if necessary. The whole is then evaporated t o about one-half its bulk the precipitated Prussian-blue is collected on a filter washed with dilute hydro- chloric acid water alcohol and ether dried a t looo and weighed. w. P. s. found to yield €rustworthy results. w. P. s. Reactions of Alanine and Glycine. L. CHELLE (Ann. Chim. and 1914 19 67-68; from Bull. SOC. pharrn. Bordeaux 1913).- Ah.nine.-About 10 mg.of the substance are dissolved in water 1 C.C. of glacial acetic acid and 0.5 C.C. of 5% sodium nitrite solu- tion are adde'd and the mixtare is boiled until nitrous vapours cease to appear; 0.1 gram of hydrazine sulphate is then added the solutJon is again boiled and 4 C.C. of concentrated sulphuric acid are added gradually. The resulting solution is divided into threeANALYTICAL CHEMISTRY. ii. 31 portions; the addition of a drop of an alcoholic guaiacol solution to one portion produces a red coloration; a yellow coloration is obtained on treating another portion with a drop of an alcoholic codeine solution whilst the third portion yields an orange colora- tion on the addition of a drop of p-cresol solution. GZycine.-The sulphuric acid solution of the diazotised substance yields a reddish- violet coloration with guaiacol a violet coloration with codeine and a green coloration with p-cresol.The reactions may be Estimation of Morphine. Lloyd’s Reagent. H. 11. GORDIN and J. KAPLAN (Amer. J. Phurm. 1914 86 461-464).- When morphine is extracted from a saturated potassium carbonate solution by shaking with a mixCure eonsisting of equal volumes of alcohol and chloroform the weight of alkaloid found is always too high (from 2 t o 15%); this method of extracting the alkaloid is therefore untrustworthy. Morphine is precipitated completely by Lloyd’s reagent but only about 90% of the alkaloid can be recovered from the precipitate by shaking this with ammonia and methyl alcohol. It is also impossible to recover the whole of the alkaloid from the precipitate formed when Strychnine salts are treated with Lloyd’s reagent ; this strychnine precipitate is tasteless but behaves physiologically like strychnine diluted with an inactive substance ; the decomposition of the precipitate cannot however be brought about by digesting it in vitro with pepsin ptyalin or F.MARINO-ZUCO and C. DUCCINI (Gccnnette 1914 44 ii 437-447).-The toxicology of e’rgot has been the subject of but few investigations which have led to contradictory results. The authors have elaborated the following method for the detection of ergot in viscera a8nd find that it gives trustworthy indications. The alcohol in which the viscera are usually consigned is separated by filtration rendered distinctly acid with tlartaric acid a8nd evaporated on a water- bath until the bulk of the alcohol is eliminated.The finely divided viscera and the sediment on the filter are heated for six hours at 7 5 O with double their volume of 95-960/ alcohol acidified with tartaric acid in a reflux apparatus; the cold alcoholic extract is filtered through new cloth and the residue firmly pressed. This extraction is repeated with fresh quantities of alcohol until the latter ceases t o become coloured the1 combined extracts being then evaporated on a water-bath t o a small volumes. The liquid is mixed with the residue from the alcohol originally present and the remaining alcohol removed by distillation at a low temperature in a vacuum. The reddish-brown turbid syrup thus obtained is dissolved in water and the solution subjected t o prolonged shaking with the twofold volume of ether in a large separating funnel; this operation is repeated with fresh quantities of ether so long as the latter becomes coloured.The ether dissolves the red colouring matter and small amounts of fat whilst the tartrates of any alkaloids present remain in the aqueous solution. The united obtained with 1 mg. of glycine. w. P. s. t r ypsin. w. P. s. Toxicological Investigation of Ergot.ii. 32 ABSTRACTS OF CHEMICAL PAPERS. ethereal extracts arc reduced t o a small volume o n a water-bath a t as low a temperature as possible and the residual liquid filtered if necessary vigorously shaken in an extractor with a small quantity oi cold saturated sodium hydrogen carbonate solution. After a long time the alkaline layer separates and exhibits a ye,llowish-red colour with a faint violet tint o r a distinct violet- red colour if the viscera contained a small or large proportion of ergot.Since this colour reaction may be masked by extraneous colouring matters the extractdon with the alkali is repeated several times and the total extract carefully acidified with concentrated hydrochloric acid and extracted with ether. The ethereal liquids which will have a more or less intense orange-yellow colour are evaporated to a small volume and the residual liquid is examined speckroscopically; in presence of the red colouring matter of ergot three absorption bands with h = 538 499 and 467 respectively are observed. Prior to the spectroscopic examination it is well to repeat the extraction with sodium hydrogen carbonate acidifica- tJon and extraction with ether a t least five o r six times in order to obtain a sufficiently pure ethereal solution of the colouring matter.The aqueous tartaric acid liquid remaining after the extraction with ether is rendered distinctly alkaline by means of saturated sodiun; carbonate solution and the liberated alkaloids dissolved by shaking several times with ether. The faintly yellow ethereal solution is concentrated Lo a m a l l volume a t as low a temperature as possible the residue being washed with a little water in which ergotinine is insoluble. The ergotinine is then either purified by Tanret’s method (A. 1878 679) or converted into its citrate by shaking the ethereal liquid several times with aqueous citric acid; the aqueous solution of the citrate is rendered distinctly alkaline by addition of sodium carbonate and the ergotinine extracted with ether. Keller’s reaction for ergotinine is next applied in the following manner. The residue left on gentle evaporation of the ether is heated slightly on a water-bath with 2-3 C.C. of 8 0.1% solution of ferric chloride in glacial acetic acid. The cold liquid filtered if necessary is placed in a testrtube and 2-3 C.C. of con- centrated sulphuric acid slowly introduced. I f ergotinine is pre- sent a blue ring appears at the surface of separation of the two liquids and the upper layer assumes a violet colour. If Keller’s reaction gives negative results the ethereal liquid which was ex- tracted with aqueous citric acid should be evaporated and the residue subjected to the same test. The above method has always been found to give certain results provided that the amount of ergot present is not less than 1 gram and that the examination of thO viscera is carried out not later than seven days after the interment of the body. Advanced putre- faction of the viscera alters the colouring matter so profoundly as to prevent its recognition. T. H. P.