THE ANALYST. 49 ORGANIC ANALYSIS. Para-nitrophenylhydrazine as a Microchemical Reagent for Acrolein and Acetone. H. Behrens. (Chem. Zezt., 1903, xxvii., 1105.)-If an aqueous solution of the hydrochloride of p-nitrophenylhydrazine is mixed with a little liquid containing acrolein, star-shaped crystalline aggregates of orange-coloured needles are formed, which, when viewed under the microscope, will be found to have a length of 150 ,u. The product should not be recrystallized before examination. In order to employ the reaction as a test for glycerin, the liquid should be evaporated with some potas- sium bisulphate and a tuft of long-fibre asbestos. By working in this way, the last portions of the water (containing some acrolein) can be distilled off without fear of frothing. The analogous reaction with aoetone is of some value for the identification uf50 THE ANALYST.denaturing materials which may contain that substance. The hydrszone obtained from acetone forms lemon-yellow crystals up to 500 ,u long, the crystals having the shape of rods with oblique ends. Acetaldehyde yields crystals of identical size and colour, but they form needles with pointed ends. This difference in appearance is so conspicuous that a denaturing agent containing acetone may be detected even in a spirit contaminated with the aldehydes of the first runnings. F. H. L. Notes on Poppy Oil. Utz. (Chew. Zed., 1903, xxvii., 1176.)-It has been recently shown by several writers (cj. ANALYST, 1902, xxvii., 363) that arachis oil is frequently contaminated with oil of sesame.Investigation of the poppy oil, which is sold for dietetic purposes as '' salad oil " proves that the same contamination occurs with this oil. Utz has examined fifteen samples of poppy oil, obtained through the German retail trade, and has observed that all give the Soltsien and the Baudoin tests for sesame. Owing to the fact that the price of sesame is higher than that of poppy (both the seeds and 88 oils), this is not a, case of intentional adulteration ; but inquiries put to manufacturers, and investigations of their methods, indicate that it is due to a want of care in the factories. Poppy and sesame seeds are usually expressed in the same plant, and the consequence is that the one oil may become oontsminated with the other. In order to ascertain what the " constants " of absolutely genuine poppy oil should be, Utz has extracted the oil by means of petroleum ether from a sample of Indian seed, light gray in colour, which contained 53.93 per cent.of oil, from a pale brown Levant (Smyrna) sample which contained 47.77 per cent., and from an undescribed sample of German seed. The iodine values of the oils, as determined by the Hubl-Waller method, were 153.48, 157.52, and 156.94 respectively. Utz's com- mercial specimens all gave figures ranging between 132 and 140-values that agree with those quoted in the books. One sample that probably contained less than 5 per cent. of sesame gave an iodine value of 151.65, another, which only showed faint indications with the Soltsien and Baudoin tests, gave 150.63.Pure poppy oil is optically inactive, as Bishop has already said ; the dextro-rotatory power of common samples is a sign of the sesame oil they contain. At 15" C. the refractive index of genuine Indian poppy oil is 1.4772, and its reading in the butter refractometer is 78.1. The corresponding figures for both Smyrna and German oils are 1.4774 and 78.4. Commercial specimens give 1.4764 and 76.7. According to Behrens, when 10 grammes of poppy oil are mixed with 10 grammes of a mixture of equal parts of nitric and sulphuric acids, a brick-red colour is produced, sesame giving a grass-green. This is not correct, for both oils, as well as several other vegetable products, yield a rapid play of colours which finally become brick-red, F. H. L. The Composition and Analysis of Linseed Oil Fatty Adds. W.Fahrion. (Zeit. mgew. Chem., 1903, xvi., 1193-1201.)-A method of separating solid and liquid fatty acids, devised by the author (Chem. Zeit., 1893, 522), was based on the oxida- tion of the unsaturated acids by means of d k a h e permanganate and the supposedTHE ANALYST. Salt. - Lithium stearate (290-38 grammes) . . . Lithium palmitate (262.38 grammes) 51 Water. I 18" C. ' 25" C. 2903.5 %32~.~ 2388-0 insolubiIity of the oxidation products in petroleum spirit. Further experiments have shown that the aolid fatty acids are also oxidized with the formation of insoluble derivatives, whilst, on the other hand, the unsaturated fatty acids yield about 2 per cent. of products soluble in petroleum spirit. These two sources of error nearly compensate one another, and the method applied to the fatty acids of linseed oil yielded (1) 8.7 per cent. and (2) 9.7 per cent.of solid fatty acids, a result in approxi- mate agreement with the results calculated from other factors. The author concludes from his experiments that in Hazura's method a large proportion of oleic acid escapes oxidation, but that the linolic, linolenic, and isolinolenic acids are almost quantita- tively attacked by the permanganate. He confirms the conclusion of Hehner and Mitchell (ANALYST, xxiii., 312), that the yield of linolenic hexabromide found by Hazura (40 per cent.) was too high, and that the precipitate contained tetrabromide, but criticises their method on the ground that the hexabromide is soluble in ether.* The melting-point of the author's hexabromide was 177" C., and he attributes the higher figure (180" to 181° C.) found by Hehner and Mitchell to too rapid heating.From these experiments and calculations drawn from tho results of other well-known methods, he concludes that the specimen of linseed oil under examination had the following approximate composition : Unsaponifiable matter, 0.8 ; palmitic and myristic acids, 8.0; oleic acid, 17.5 ; linolic acid, 26.0 ; linolenic acid, 10.0 ; isolino- lenic acid, 33-5; and glycerin radicle (C,H,), 4-2 per cent. C. A. M. New Methods of Separating Fatty Acids. A. Partheil and F. Ferie. (Archiv der Pharrn., 1903, ccxli., 545-570.)- It is shown that in Farnsteiner's method of separating solid and liquid fatty acids by treatment of their lead salts with benzene (ANALYST, xxiii., 285) part of the oleic acid i s precipitated in the form of a double stearate and oleate of lead.To obviate this, experiments were made with different monovalent metals, of which lithium was found to be the most suitable, whilst rubidium acetate only effected a partial precipitation of stearic and palmitic acids from an alcoholic solution. The solubility of the different lithium salts in water and alcohol is shown in the following table, which gives the number of litres required to dissolve the different molecular weights in grammes : Alcohol. Specific Gravity 0.797. 18" C. 708.3 329-5 127.4 49-3 31-74 25" c. 545.6 274.5 110.5 46.62 I 28.57 The method eventually adopted was to dissolve 0.25 gramme of the fatty acid in 50 C.C.of absolute alcohol, to neutralize the solution with alcoholic potassium * NOTE BY AnsTRACTOR.-specia~ stress was laid on this point by Hehner and Mitchell, and a means of obviating the error devised.52 THE ANALYST. hydroxide, and after dilution with 50 O.C. of water to add an excess of a, 10 per cent. alcoholic solution of lithium acetate. The precipitate was collected on a weighed filter, washed with 50 per cent. alcohol, dried, and weighed. I n this way the following results were obtained with pure fatty acids : Found. Gramme. Gramme. Lithium stearate . . . ... 0-2552 0.2495 Lithium palmitate . . . ... 0.2558 0.2525 Lithium myristate . . . ... 0.2565 0.2347 No trace of precipitate was given by the lithium laurste or oleate. The author therefore concludes that stearic and palmitic acids are quantitatively precipitated and myristic acid nearly so in this method, For the separation of the lauric acid and residual myristic acid in the solution, Farnsteiner's method (Zoc.cit.) can be employed, for the author has proved that these acids, unlike st-earic and palmitic acids, are precipitated f-rom a benzene solution as lead salts quite free from lead oleate. I n calculating the composition of the solid fatty acids separated by these methods, they were converted into barium salts and their molecular equivalent determined (cf. ANALYST, xxi., 318). The fatty acids recovered from the benzene solution of the lead s d t s were dissolved in alcohol, and the solution neutralized and treated with a 10 per cent. alcoholic solution of barium acetate.The barium salts were dried and extracted with ether containing some water, which was found to dissolve the barium salts of the more unsaturated fatty acids. The following results, agreeing well in duplicate determinations, were obtained by these combined methods with the mixed fatty acids of different fats : Calculated. Fat. Butter, A _.. ... ,, B ... . . I Margarine . . . ... American lard (iodine . . . Human fat, A ... ? ? €3 ... value = 65.8) Stearic Acid. Per Cent. 6-54 10.33 19.75 8.16 12-30 12.46 Palmitic Acid. Per Cent. 17.9 14-23 6.22 4.36 29.25 27.12 Myristic Acid. Per Cent. 10.65 12.25 13.72 14.03 None found. ?, Lauric Acid. Per Cent. 17.08 14.44 6.83 13.08 None found. ? 9 Unaatur- $ted Acids Per Cent. 30.08 33.03 46.91 53.75 49.07 52.93 Highly Un- saturated Acids in Unsaturated Acids.Per Cent. 5.40 4-15 20.3 10.03 - - Seven specimens of human fat of different origin gave the following resuIts : Iodine value, 57-21 to 66.30 ; Reichert-Meissl value, 1.38 to 2-19 ; saponification value, 194.2 to 198.1; and Hehner value, 93.92 to 96.0 (cf. ANALYST, xxi., 171). C. A. M.THE ANALYST, 53 Pheephomolybdic Acid as a Reagent for the Identification of Vegetable Oils. F. Seiler and A. Verda. (Chem. Zeit., 1903, xxvii., ll2l.)-The authors hold that the Welmann reaction is not capable of distinguishing vegetable fats with certainty from oils of animal or mineral origin. I t does not seem that alkaloids or glucosides in the oils are the sources of the colours which vegetable materials give ; indeed, as Welmann himself has admitted, the quantity of alkaloidal matter in such vegetable oils is so small that other reagents (which are not lacking in delicacy) fail to show it.Amides of the aliphatic and aromatic series (e.g., anilides, toluides, and naphtha- lides) give no precipitates with phosphomolybdic acid. Derivatives in which an acid radicle has entered the NH, group (urea, asparagine, etc.) give neither colour nor pre- cipitate. Amines yield precipitates and a dyestuff which varies in colour from blue to green; on adding ammonia, the precipitate generally dissolves, and the colour changes to a pure blue. F. H. L. Determination of Caoutchouc in Rubber Goods. C. 0. Weber. (Ber., 1903, xxxvi., 3103 ; through Chem. Zeit. Rep., 1903, 3OO.)-For the purpose of estimating the amount of true rubber in rubber articles, the author prepares the addition com- pound of caoutchouc with nitrogen dioxide, a body having the empirical formula C,,H,,N,O,, and containing 59.65 per cent.of its organic constituent. A weighed quantity of the sample is dissolved in benzene. Nitrogen dioxide is prepared by heating lead nitrate, about 20 grammes being needed for each analysis. The gas is passed through a drying cylinder charged with glacial phosphoric acid, and led into the solution until its colour becomes red-brown. The product is allowed to rest for an hour, and the benzene is run off through a filter, any solid matter being returned to the flask, which is then laid on its side, and dried at a temperature of 50" C. The dry residue is dissolved in acetone, and after a short time the gray solid matter (consisting of the mineral matters and albuminous constituents of the sample) which separates out of the deep yellow solution is collected on a tared filter, washed with warm acetone, dried, and weighed.The filtrate and washings are poured into eight times their volume of water, when the pure caoutchouc derivative is precipi- tated as a somewhat flocculent yellow substance. This is brought on to a tared paper, washed with lukewarm water, dried at a temperature not exceeding 90" C., and weighed. I t may be considered to contain 60 per cent. of caoutchouc. In the original article Weber gives a method for the examination of vulcanized rubber, and for the estimation of the various substancas it contains.F. H. L. Estimation of Nitrogen in Creatine by the Kjeldahl Process. C. Beger, G. Fingerling, and A. Morgan. (Zeits. Physiol. Chem., 1903, xxxix., 329 and 467; through Chem. Zeit. Rep., 1903, 282,)-Kutscher and Steudel have stated that the nitrogen in creatine, creatinine, uric acid, lysin, and histidin cannot be estimated by the Kjeldahl process. This, however, is quite incorrect so far as creatine is concerned, and the authors of this paper therefore assume it to be equally incorrect54 THE ANALYST. in the case of the other substances mentioned, since, when employing the method prescribed by the Verband der landwirtschaftlichen Versuchsstationen, they alwayB obtain perfectly accurate results. Malfatti (loc. cit., 282) agrees with Beger, Fingerling, and Morgan, more espe- cially in the case of the compound of creatine with zinc chloride.He heats the substance with sulphuric acid, but without copper sulphate or mercury, till he obtains a brown liquid that boils sm9othly. He then cools it, adds a sufficient quantity of potassium permanganate in solution, and heats again till the water is driven off and the residue is colourless. For all purposes Malfatti prefers to use the permanganate as a solution rather than as a solid, even when mercury or copper sulphate is also employed. F. H. L. The Behaviour of Methyl Violet and Tropceolin with Certain Acids. Schumacher-Kopp. (Chem. Zeit., 1903, xxvii., 1176.)-It is usually stated in text- books that methyl violet changes to a green or blue in presence of mineral.acids.This is true for hydrochloric, sulphuric, nitric, and phosphoric acids, but not for boric acid, which produces no effect. The colour of methyl violet is changed to blusgreen by oxalic, tartaric, and lactic acids, to blue by citric acid, while acetic acid is perfectly inert. The colour of tropEolin is altered to red-violet by hydrochloric, sulphuric, nitric, and phosphoric acids ; like methyl violet, it is not affected by boric acid. Oxalic and tartaric acids turn it red-violet, citric acid orange, lactic acid rose colour, and acetic acid cherry red. F. H. L. A New Indicator obtained from Meta-toluidine. J. Troger and W. Hille. (J. Prakt. Chem., 1903, lxviii., 297 ; through Chem. Zed. Rep., 1903, 282.)-Meta- toluidine is dissolved in sulphuric or hydrochloric acid, the solution is diazotized, thoroughly cooled, and either treated with a current of sulphur di-oxide or mixed with an aqueous solution thereof. The red product is purified by boiling it with water, after which it is treated with 8 warm, strong, aqueous solution of potassium acetate, and the warming is continued till a dark yellow potassium compound is obtained. This is converted into the corresponding sulphonic acid, which crystallizes in ruby red prismatic needles having the formula HS0,.C7H,.N:N.C7H,(NH,),. The alkali salts of this acid form indicators which in colour and general properties resemble helianthin, but they are very much more sensitive. F. H. L.