130 THE ANALYST. ORGANIC ANALYSIS. Blackberry-seed Oil. R. Krzizan. (Chem. 12cv. Fett- u. Hurx-Id., 1908, 15, 7-9, 29-30.)-By extraction of the crushed seeds with petroleum spirit, 12.6 per cent. of a greenish-yellow oil with a red fluorescence, and possessing a characteristic odour, was obtained. The fatty acids were fractionated into 4.7 per cent. of solid fatty acids, mainly palmitic acid, and 95.3 per cent. of liquid fatty acids containing about 80 per cent. of linolic acid, 1.5 per cent. of linolenic acid, and 1-5 per cent. of isolinolenic acid. The tetrahydroxystearic acid obtained by oxidising the liquid fatty acids with alkaline permanganate melted at 161" C., instead of the usual 173Oto 174" C. The oil itself absorbed 6.87 per cent. of oxygen in three days in Livache's test, and dried in two to three days in the film test.It gave the following values : Specific gravity, 0.9256 ; acid value; 2.03 ; saponification value, 189.5 ; Reichert-Meissl value, 0.0 ; Hehner value, 96.3 ; iodine value, 147.8 ; and unsaponifi- able matter, 0.83 per cent. Fatty Acids.-Specific gravity, 0.9070 ; saponification value, 199.9 ; iodine value, 155 ; acetyl value, 13.9 ; and mean molecular weight, 280.9. The phytosterol (0.6 per cent.) separated from the oil melted a t 133" C . Neither Halphen's nor Baudouin's colour reaction was given by the oil. C. A. M. The Polarimetric Determination of Cane Sugar. Effect of Basic Lead Acetate on the Optical Activity and Copper-Reducing Power of Sugar Solutions. F. Watts and H. A. Tempany. ( J o z m . SOC.Chem. Ind., 1908, 27, 33-57.)-The authors show that excessive amounts of basic lead acetate exercise anTHE ANALYST. 131 appreciable effect on the optical activity and reducing power of solutions of invert sugar. When excess is avoided, and it is easily avoided in practice, clarification by dry anhydrous basic lead acetate involves no appreciable error. In raw-cane juices a slight excess of basic lead acetate produces a greenish-yellow coloration which may serve as an indicator for the process of Clarification. The reduction of the optical activity of solutions of invert sugar by basic lead acetate is due to the forma- tion of a soluble lead Itmulosate, which can only exist in presence of the basic group of the lead acetate. Low-grade products, such as molasses, can only be clarified by basic lead acetate solution if a large excess of the reagent be employed, and in such cases the errors involved by the volume of the precipitate and the action of the excess of lead on the locvulose are large.On the other hand, dry basic lead acetate only effects a partial clarification of such products. The authors prefer to add the dry basic lead acetate to a solution of twice the normal weight of molasses per 100 C.C. and filter. Of the filtrate, which is still dark, 50 C.C. are saturated with sulphur dioxide, diluted to 100 c.c., and filtered for polarisation. The use of animal charcoal is quite inadmissible. Clarified juices always contain small quantities of lead salts, but these are not sufficiently large in careful working to introduce an appreciable error in the determination of reducing sugars by Fehling’s solution. Direct experiments with solutions of pure invert sugar, containing about 0.5 gram per 100 c.c., showed that 1 C.C.of basic lead acetate solution caused the apparent removal of 0.006 gram of invert sugar, but since the total quantity of lead solution required for clarification rarely exceeds 2 c.c., there is little danger of an excess of 1 C.O. ever being present. J. F. B. A Colour Reaction for Carbohydrates and its Relation to their Molecular Weights. L. Wacker. (Ber. dczit. Ghem. Ges., 1908, 41, 266-275.) --p-Phenyl- hydrazine sulphonic acid condenses with aldehydes and alcohols of the aliphatic series, in presence of air and of alkali hydroxide in excess, to form red colouring matters.The shade of these colouring matters is approximately constant, so that their intensities are comparable and show the following regularities : The intensity of the colours of equimolecular quantities is constant ; i.e., the intensity due to equal quantities decreases with increased molecular weight ; the rapidity of the reaction is inversely proportional to the molecular weight, and is also proportional to the concentration. The simplest aldehydes, such as formic and acetic aldehydes, are the most sensitive ; they react a t once, and may be detected at concentrations of -zaYGa. Polyvalent alcohols and carbohydrates are also very sensitive, and react distinctly at concentra- tions of a&v; in this group the reaction is delayed for a longer or shorter time, according to the molecular weight, and begins to appear after about fifteen minutes.With the primary monovalent alcohols the limit of sensibility is %&. At higher concentrations colorations are obtained with substances such as acetone, lactic acid, citric acid, urine, and proteins. For quantitative determinations, separate portions of 0.25 gram of phenylhydrazine sulphonic acid are weighed out into a series of wide- mouthed bottles (200 grams) of equal dimensions. The substances to be tested, dissolved or suspended in 100 C.C. of water, are then added, and followed imme- diately by 15 C.C. of a 33 per cent. solut.ion of sodium hydroxide to each bottle.132 THE ANALYST. Each series is accompanied by a blank from which the substance, alcohol or alde- hyde, is omitted.The colorations produced are estimated comparatively in colori- meter tubes at the end of four, five, ten, twelve, or twenty hours. The blank should show only a yellow coloration, but it may absorb formaldehyde or other vapours from the air which would interfere with the test. The method, besides being applicable for the qualitative and quantitative estimation of many substances of this class, may also be used with considerable success for the estimation of the molecular weights or number of hexose groups in polysaccharides. As standards for this test, the disaccharides, maltose and lactose, are preferable to the simple hexoses which give colours with a bluish tone: The test may be made in two ways-either direct or after inversion.When tested direct, various multiple weights of the polysaccharide are compared with or a ~ \ c solutions of maltose until equivalent colours are produced; By the inversion method, a portion of the fully inverted polysaccharide is compared with different quantities of the original polysaccharide, and the ratio between the two solutions which match gives the ‘‘ inversion quotient.” By both of these methods it W ~ E found that soluble starch is equivalent to ordinary starch (which, however, had undergone the usual change in contact with the alkali hydroxide of the test solution). These starches are equivalent to three maltose groups, and by the inversion method they show a quotient of six. Erythrodextrin is equivalent to two maltose groups, and has an inversion quotient of four.Glycogen appears to be equivalent to at least two molecules of soluble starch. J. F. B. Estimation of Carbon Bisulphide in Benzene. J. Bay. (Comptes ILeitd.. 1908, 146, 132.)-The method of Liebermann and Seyewetz for the detection of traces of carbon bisulphide in benzene by means of phenylhydrazine may also be used for its quantitative estimation. The crystalline precipitate which phenyl- hydrazine gives with carbon bisulphide, CS,(C,H,.NH.NH,),, is very unstable in solution, but suficiently stable in the dry state for an estimation to be made. Precipitation is complete in two to three hours, and the precipitate is collected on double counterpoised filter-papers, washed with benzene till free from the excess of phenylhydrazine, and dried in VUCZLO.The results are slightly too high, owing to the difficulty of washing the precipitate completely. Thus, in typical test analyses,, in which 1963 and 12.630 per cent. of carbon bisnlphide were added to benzene, the amounts found were 1-269 and 12.640 per cent. respectively. C. A. Ill. Volumetric Estimation of Basic Dyestuffs by Means of Acid Dyestuffs. L. Pelet and Garuti. (Zeit. Fa& Ind., 1908,7,44-45.)-The methods recommended for the estimation of basic dyestuffs comprise: (1) Titration of methylene blue with Crystal Ponceau, Carmine in sodium carbonate solution, Pyramine orange (B. A. S.F.) and Cotton brown (B.A.S.F.). (2) Titration of Safranine with Helvetia blue, Sapthol yellow, S, and Acid violet, 6BN. (3) Titration of Fuchsin and New Fuchsin with Helvetia blue, Alkali blue, and Light green SF, bluish.(4) Titration of Auramine 0 and Auramine G with Congo red, Acid violet, 5BN, Oxamine red, Alkali blue, and Violet black. (6) Titration of Vesuvine with Violet black, Ocelline, Chromazone red, and water soluble Fast blue R. (5) Titration of Chrysoidine with Helvetia blue.THE ANALYST. 133 The methods all depend upon the precipitation of the basic dyestuE and deter- inination of the end-point by ‘‘ spotting ” on filter-paper. Solutions containing 1 to 3 per cent. of known pure dyestuffs have given very accurate results in this way, the precipitates formed being pseudo-salts of definite and constant composition. The end-point of the reaction may also be found by determining the electrical conduc- tivity of the solution, which is decreased by the addition of acid dyestuffs, and reaches the minimum when precipitation is complete.C. A. M. Test for Cholesterol. J. Lifschutz. (Ber. ricut. C‘hem. Gcs., 1908, 41, 252-255.)-The oxidation products of cholesterol, oxycholesterol ester and oxy- cholesterol, may be made to serve as tests for the presence of this alcohol. The oxidation is best carried out in acetic acid .solution by means of a peroxide of an organic acid radical. A few milligrams of cholesterol are dissolved in 2 to 3 C.C. of ,glacial acetic acid, and a little benzoyl peroxide added. The solution is boiled up once or twice, and 4 drops of strong sulphuric acid are added when it is cold. The mixture then assumes a violet-blue or green colour, according to the quantity of peroxide employed ; the first oxidation product giving the blue and the second the green coloration.The green coloration shows a characteristic absorption-band in the middle of the red portion of the spectrum between the lines C and d. Oxycholes- terol itself frequently occurs in the internal organs of animals. J . F. B. New Tests for Cholesterol and Oxyeholesterol. L. Golodetz. (Chem. hit., 19013, 32, 160.)-A new test for cholesterol is described which possesses the advantage that it may be applied to solid preparations of tissues under the micro- scope. The reagent consists of a mixture of 5 parts of concentrated sulphuric acid and 3 parts of formaldehyde solution; 1 or 2 drops of this reagent colour solid cholesterol a blackish-brown. It reacts only with free cholesterol, and may be used for the detection of the free alcohol in preparations of cholesterol esters.Another reaction, which, however, cannot be used as a micro-chemical test, consists in adding to a few drops of a solution of cholesterol in trichloracetic acid 1 drop of 30 per cent. formaldehyde solution, which produces an intense blue coloration. A test for oxycholesterol, the oxidation product of cholesterol, first isolated by Lifschiitz from wool fat, and also observed in bone and blood fats, consists in bringing the substance in contact with a few drops of liquefied trichloracetic acid. Oxycholesterol gives a, green solution showing a dark bend in the red portion of the spectrum. J. F. B. Analysis of Mixtures of Volatile Fatty Acids. A. Lasserre.(Ayzn. de L’Instit. Pastew, 1907, 21, 829 ; through Chm. Zeit. Rep., 1908, 32, 67.)-Acetic and formic acids may be separated from butyric and valeric acids by extracting the aqueous solution of the mixed acids with benzene, in which only the last-named acid dissolves, and from which they may be recovered by treatment with baryts solution. A. G. L. New Reactions for the Characterisation of Mercerised Cotton. d. Hubner, (Jown: SOC. Chem. Ind., 1908, 27, 105-lll.)-The author has standard-134 THE ANALYST. ised the well-known colour reactions between hydrated cellulose and solutions of iodine in presence of various salts, and has deduced therefrom a means by which mercerised cotton may be identified, and the strength of the sodium hydroxide which had been used in the mercerising process approximately determined.The method is based on the observations that mercerised cotton possesses a greater affinity for iodine than ordinary cotton, and that cotton mercerised under tension absorbs less iodine than cotton similarly mercerised without tension. Bleached cotton absorbs less iodine than unbleached cotton. I n carrying out the test, a series of samples of cotton fabric mercerised (under tension) with solutions of sodium hydroxide of various known concentrations must be prepared. If the sample to be tested be coloured, either the colour must be etripped before applying the reagent, or the test standards must be dyed to the same shade with the same dyestuff. ( a ) Iodine-Potassium Iodide Reagent.-The reagent contains 30 grams of iodine dissolved in 100 C.C.of a saturated aqueous solution of potassium iodide. If samples of cotton which have been mercerised with solutions of sodium hydroxide of different strengths be immersed in this reagent, and then washed with wclter, the black- brown coloration which they have acquired disappears more slowly the stronger the mercerising solution, up to R limit of 60" Tw. Unmercerised cotton loses its colour very rapidly. ( b ) Iodi?lc-Zi?zc Chloride Reagent.-The zinc chloride solution must be accurately prepared so as to contain 280 grams of zinc chloride in 300 C.C. of the solution. To 100 C.C. o€ this solution are added either 20, 10, or 5 drops of a solution of 1 gram of iodine and 20 grams of potassium iodide in 100 C.C. of water. The reagents containing 20 and 10 drops of iodine solution are suitable for determining the strength of the mercerising liquid a t concentrations corresponding to 10" to 30' Tw., whilst the reagent containing 5 drops is employed for the recognition of products mercerised with liquors of 30" to 60' Tw.In using the zinc chloride reagents, the colorations of the sample and the test standards are compared directly, washing, as in the case of the potassium iodide reagent, being unnecessary. If desired, the rate at which the colours fade on exposure to the air may also be determined. In applying the test to fabrics, the samples should first be wetted with water and pressed between filter-paper. In testing coloured samples, a piece may be immersed in the zinc chloride solution without iodine, to serve as a blank for comparison. Dark colours and colours dyed on tannin mordants should b3 stripped before testing (cf.ANALYST, 1908, 55). J. F. B. The test may be used in one of two forms : Starch, if present, must be completely removed. Determination of Phenols in Gas Liquors. F. W. Skirrow. (Journ. SOC. Chcm. Ind., 1908, 27, 58-62.)-For the estimation of phenols in gas liquors by the iodometric method, the liquors must first be freed from sulphides and cyanides. For this purpose 100 C.C. of the filtered gas liquor are treated with an excess of ammonium polysulphide, to convert the cyanides into thiocyanates. The mixture is allowed to stand, and is then diluted to 200 C.C. The sulphides are precipitated byTHE ANALYST. 135 the addition of lead carbonate, and the precipitate is filtered off.An excess of sodium hydroxide (25 C.C. of a 50 per cent. solution) is then added to 100 C.C. of the filtrate, and the solution is evaporated until the salts begin to crystallise out. The residue is washed into a distillation flask of 1 litre capacity and diluted to about 150 c.c., and when cold is acidified with sulphuric acid. The liquid is distilled until the salts begin to crystallise out-thus (‘ distillate No. 1 ” is obtained ; 100 C.C. of water are then added to the residue in the flask, and ( ( distillate No. 2 ” is collected in a separate receiver. The whole of the phenol is contained in the three distillates. Each of these is neutralised separately with calcium carbonate and a little lead carbonate, and redistilled in the same order as they were first collected. The final distillates are mixed and diluted to 500 C.C.I n titrating the phenols, the quantity of sodium hydroxide present is of some importance. I t is best to perform a preliminary titration in presence of a, large excess of sodium hydroxide, and to use finally the proportions of 4 molecules of sodium hydroxide to 1 molecule of phenol found at first. The solution is made alkaline with this quantity of sodium hydroxide ; it is then warmed to 60’ C., and treated with excess of TG iodine solution. The flask is immediately closed, and the contents are allowed to cool; they are then acidified, and the excess of iodine is titrated back with :v thiosulphate. One molecule of phenol consumes 6 atoms of iodine.The author has ascertained that the presence of sulphites, thiosulphates, and thiocyanates has no influence on the results. Since cresols are present in the gas liquor as well a8 phenol, the results are about 5 per cent. too law on this account. The liquid after titration has a rose-pink coloration, owing to the presence of tri-iodophenol. This coloration map be made the basis of a colorimetric method for the estimation of small quantities of phenol which cannot be titrated with accuracy. J . F. B. I n a similar manner a, third distillate is obtained. Estimation of Picric Acid. M. Busch and G. Blume. (Zeit. a y c w . Chena., 1908, 31, 354-355.)-The method proposed is based on the insolubility of ‘( nitron ” picrate, a solution containing 1 part of picric acid in 250,000 parts of water giving a distinct precipitate when treated with nitron ” acetate solution (cf.ANALYST, 1907, 32, 349). The estimation is carried out as follows : The picrate solution, containing about 0.15 gram of the acid in 150 C.C. of water, is acidified with 2 C.C. of dilute sulphuric acid, and heated just to boiling. Ten C.C. of icnitron” acetate solution (10 per cent. of ‘( nitron” in 5 per cent. acetic acid) are added slowly, and the mixture is allowed to cool to the ordinary temperature. The precipitate is collected on a weighed filter, washed with from 50 to 100 C.C. of cold water, dried for one hour at a temperature of l l O o C., and weighed. The weight of the ‘( nitron ” picrate- C,,H,GN,.C,H3N307-multiplied by the factor 0.42529, gives the weight of the picric acid. The method is applicable to the estimation of picric acid in mixtures of various high nitrated phenols, pyridin picrate, quinolin picrate, etc., and the above- mentioned limits as to dilution may be varied considerably without influencing the accuracy of the results. The solution, however, must not contain bromides, iodides, chlorates, perchlorates, nitrates, nitrites, or chromates. w. P. s.136 THE ANALYST. Examination of Salicylic Aldehyde for Use in Testing for Fuse1 Oil. H. Kreis. (Cliem. Zcit., 1908, 32, 149.)-The author warns against the use of impure salicylic aldehyde, which may give a red colour even with alcohol free from fusel oil. The aldehyde may be tested by adding 0.5 C.C. of its 1 per cent. solution and 10 C.C. of concentrated sulphuric acid to 5 C.C. of 95 per cent. alcohol known to be free from fusel oil. After cooling, the solution should be yellow in colour, resem- bling picric acid, but not reddish. Or 3 drops of the aldehyde are mixed with 10 drops of sulphuric acid. If the aldehyde is pure, a light orange-red colour will be obtained, impure samples giving brownish-red colours. On adding 5 C.C. of 95 per cent. alcohol to the liquid, it will become colourless if the aldehyde was pure, whilst in the contrary case the solution will turn red (cf. ANALYST, 1903, 28, 293). -1. G. L.