NOVEMBER, 1916. Vol. XL., No. 476. THE ANALYST. ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOOD AND DRUGS ANALYSIS. New Silicotungstic Acid Method for the Estimation of Alkaloids. A. Ferencz and L. David. (Pharm. Post., 1914, 47, 559-563; through J. Chem. SOC., 1915, 108, ii., 600.)-Ten C.C. of a solution containing the salt of the alkaloid are treated with 5 C.C. of 10 per cent. silicotungstic acid solution; the precipitate is collected on a filter, and washed with 10 C.C.of very dilute hydrochloric acid. The precipitate is then transferred to a flask, and shaken for five minutes with 5 C.C. of water and 10 C.C. of 20 per cent. sodium hydroxide solution ; after the addition of 15 grms. of sodium chloride and 100 C.C. of ether (in the case of nicotine, 50 C.C.of a mixture of ether and light petroleum are used), the mixture is shaken for ten minutes, allowed to separate, and one-half of the ethereal solution is drawn off and titrated with & hydrochloric acid, using iodoeosin as indicator. Estimation of Sulphates in Bread. C. D. Howard. (J. Ind. and Eng. Chem., 1915, 7, 807.)-O’Sullivan (ANALYST, 1914, 39, 425) and Cripps and Wright (ibid., 429) have shown that sulphuric anhydride tends to be lost under certain conditions when organic matter is destroyed by incineration.The latter authors and also Elsdon (ibid., 1915, 142) have described methods depending on extraction of the bread, flour, and the like, with acids without previous incineration. The method now described differs from that of Elsdon in the use of dilute instead of con- centrated hydrochloric acid, which the author has found to give rise bo trouble, due to progressive precipitation of protein substances.The air-dried bread (25 grms.) is digested with 200 C.C. of 5 per cent. hydrochloric acid for three or four hours on the water-bath at a gentle heat,” the cooled extract is filtered, the residue washed with warm water, and sulphate in the filtrate is precipitated by means of barium chloride.As a rule the precipitate is clean ; when otherwise, as shown by charring on ignition, the ignited precipitate is extracted on the Gooch crucible with acidulated hot water and reignited. G. C. J. Estimation of Fat in Ice Cream by the Babcock Method. C. A. A. Utt. (J. Ind. and Eng. Chem., 1915, 7, 773.)--The hydrochloric-aoetic acid method is not satisfactory if much gum thickener, gelatin, or condensed milk has been added, a,462 ABSTRACTS OF CHEMICAL PAPERS white precipitate tending to risein theneck of the bottle and obscure the fat column.A mixture in equal parts by volume of sulphuric acid (sp. gr. 1.83) and glacial acetic acid (99-5 per cent.) gives excellent results.The sample is warmed to 40' C., and, if the fat has separated, a little powdered sodium hydroxide is added to aid emulsification. Of the well-mixed sample, 9 grms. are weighed into a 10 per cent, Babcockmilk bottle, 12 to 15 C.C. of the acid mixture are added, the whole shaken, and then heated in a, hot water-bath, with occasional shaking, until of a dark chocolate colour. The bottles are centrifuged for ten minutes, shaken again, filled to the neck with hot water, the contents mixed, and centrifuged again for three minutes.Boiling water is added to bring the fat column into the graduated neck, and the tubes are centrifuged again for two minutes. The author reads from the bottom of the fat to the extreme top of the meniscus, The Babcock reading, multiplied by two, gives the percentage of fat in the sample.The extreme error of the method, with ice cream of 12 to 17 per cent. fat content, is 0.15 per cent. on the sample. G. C. J. Estimation of Hgdrastine in Plants and in Extracts of EVdrast8s Cunud ensis by the Methods prescribed in the Various Pharmacopaeias, and a New Method for the Estimation of Berberine in Extracts. L. David.(Pharm. Post., 1915, 48, 1-4; through J. Chem. Soc., 1915, 108, ii., 601.)-The author has determined the solubility of hydrastine in various solvents; 1 grm. of the substance is dissolved by 227.27 grms. of ether at 20° C., by 22727027 grms. of light petroleum (boiling-point 40° to 60' C.) at 15' C., by 11627.88 grms. of light petroleum (boiling-point 50" to 85" C.) at 15* C., and by 1791 grms.of benzene at 15' C. Hydrastine is soluble in chloroform. The methods given in the German and Belgian pharmacopoeias are the most trustworthy for the estimation of hydras- tine ; then follows that of the Austrian pharmacopoeia, whilst the French, Dutch, Swiss, and U.S.A. methods are untrustworthy, as the hydrastine obtained by their use is very impure. The U.S.A. method may be improved by removing the alcohol before the hydrastine is extracted with ether.I n the estimation of hydrastine, alcohol must be removed previously, berberine should be separated by precipitation with potassium iodide or hydrochloric acid, and the ethereal extract must be mixed with light petroleum, and the mixed solution treated with tragacanth. A method proposed for the estimation of berberine in hydrastine extracts depends on the precipitation of berberine and hydrastine by potassium bismuth iodide, and the separation of the hydrastine by extraction with ethyl acetate.The precipitate is then treated with 10 per cent. sodium hydroxide solution, and the berberine sxtracted with a mixture of equal parts of ether and chloroform. Estimation of the Quantity of Meat Extract in Soup Tablets.T. Sudendorf and 0. Lahrmann. (Zeitsch. Urtters. Nahr. Genussm., 1914, 29, 1; through J. SOC. Chm. Ind., l915,34,1027.)-Ten grms. of the sample are dissolved in water, the solution diluted to 100 c.c., filtered, and 20 C.C. of the filtrate is mixed with 10 C.C. of 1 hydrochloric acid and the mixture evaporated to dryness. The brown-black residue is treated with water, neutrslised with alkali solution, d o gFOOD AND DRUGS ANALYSIS 463 litmus as indicator, then diluted to about 75 c.c., and 1 per cent. permanganate solu- tion containing 2.5 per cent.of sodium chloride is added, drop by drop, until a slight excess is present. The excess of permanganate is destroyed by the addition of 3 per cent. hydrogen peroxide solution containing 1 per cent.of glacial acetic acid, the mixture heated for ten minutes on a water-bath, filtered, and the filter washed with hot water. The filtrate is evaporated to a small volume, transferred to a 50 C.C. flask, diluted to about 20 c.c., 10 C.C. of 10 per cent. sodium hydroxide solution and 20 C.C. of saturated picric acid solution are added, the mixture is diluted to 50 C.C.after the lapse of five minutes, and the coloration compared with that of a standard dichromate solution in the usual way. (See ANALYST, 1909, 34, 475.) Oil of Pistncia Lentiscus. E. Sernagiotto and N. Vita. (,d?zn. C h h . anal., 1915, 4, 92-93.)-The cold-drawn oil, freed from substances volatile with steam, ha.d the following characteristics : Sp. gr. at 15' C., 0.9188 ; butyrorefracto- meter reading (Zeiss), 66.0 ; acid value, 15.92 ; saponification value, 165.58 ; saponification value after acetyl'ation, 203.03 ; Hehner value, 94-82 ; iodine value, 83.61 ; and iodine value of fatty acids, 85.96.The fatty acids consisted of palmitic and oleic acids. C. A. M. Theory of Wetting Power and the Determination of the Wetting Power of Dipping and Spraying Fluids containing a Soap Basis. W.F* Cooper and W. H. Nuttall. (J. Agric. Sci., 1915, 7, 219-239).--It is recognised that the efficiency of sheep-dips or agricultural sprays does not depend solely on the strength of the toxic constituent, but that its ability to wet and spread over the greasy surface to be treated is a matter of the greatest importance. This wetting power cannot be estimated by chemical analysis, and the factors which operate form the subject of a lengthy theoretical discussion in the present paper.To a, certain extent the wetting power of a liquid towards a solid depends on the condition that the surface tension of the solid be greater than the sum of the surface tension of the liquid and the interfacial tension. This latter is a function which comes into play, for example, when a drop of oil is suspended in dilute alcohol of the same density, and is the tension existing at the interface between the two immiscible liquids.A liquid having a good wetting power should have a low surface tension and a low interfacial tension in relation to a greasy or waxy surface, but the latter tension is the more important in determining the wetting power, since a low inter- facial tension enables the liquid to spread over the whole surface, even if its own surface tension be fairly high.But the general problem is complicated by secondary factors-viz. , the solvent properties of the wetting liquid, which ensure actual contact between the liquid and the solid, and the phenomenon known as surface concentration (capillary adsorption).This last has a curious effect on the physical surface properties of liquids containing saponin or gdatin, causing them to '' run " very readily, forming liquid planes the high surface viscosity of which prevents rupture and formation of globules. On account of these and other complications,464 ABSTRACTS OF CHEMICAL PAPERS no perfectly general and satisfactory method for the determination of wetting power has yet been evolved, but in the case of liquids prepared with a soap basis, the surface tensions of which are substantially constant, comparative valuations may be obtained in a simple manner by an estimation of the interfaqial tensions towards a thick neutral oil (liquid vaseline) taken to represent the greasy solid to be treated.The method employed is a modification of Donnan’s drop-pipette method. The standard oil is run from a pipette, the end of which is bent round so as to point upwards, immersed in a column of aqueous liquid containing the fluid to be tested at a concentration similar to that recommended for practical use. For com- parative purposes the interfacial tension may be taken as inversely proportional to the drop number, and the wetting power as directly proportional to the same.The aperture of the pipette is somewhat restricted, and ground perfectly level ; the walls of the aperture should be fairly thin. The portion of the pipette above the bulb has a side branch for filling by suction, and both side branch and upper vertical limb are controlled by stopcocks; in addition, a piece of fine capillary tubing is attached to the top of the pipette to regulate the flow of oil so as to allow about ten seconds between the drops when immersed in water.The capacity of the pipette may be 25 c.c., and the diameter of the aperture 2.7 mm. After adjusting the charged pipette in a cylinder containing the liquid to be tested, the stopcock is opened and the number of drops counted as the oil passes from the upper mark to the lower mark.If streaming occurs, as with liquids of very low interfacial tensions, a pipette with smaller orifice must be used. The value of the standard vaseline should be determined against distilled water at 20’ C., and the wetting power of the solution expressed as the ratio of its drop number to that of pure water x100.Quite useful results may be obtained by this method with most of the commercial fluids, but it must only be applied to soap solutions, and is valueless for saponin and similar preparations. J. F. B. Occurrence and Estimationof Citric Acid in Wine. (R. Kuna. Arch. Chem. Mikrosk., 1914 ; through J. Chzem. SOC., 1915, 108, ii., 595-596.)-Stahr’s reaction is recommended for the detection of citric acid; the reaction depends on the formation of a white precipitate of pentabromacetone, when bromine and potas- sium permanganate act on citric acid.The author prefers to use potassium bromide in place of bromine ; 10 C.C. of the solution to be tested are treated with 1 C.C. of sulphuric acid (1 : 1) and 0.3 C.C. of 37.5 per cent.potassium bromide solution; 1 C.C. of 5 per cent. permanganate solution is then added, the mix- ture heated for five minutes at 45’ C., and any traces of manganese dioxide are removed by addition of a drop of ferrous sulphate solution containing sulphurio acid. A distinct turbidity is obtained with 1 mgrm. of citric acid. Wines usually give a precipitate with bromine done, and in this case the test is modified by evaporating 50 C.C.of the wine to 10 c.c., adding 2 C.C. of dilute sulphuric acid, then 10 C.C. or more of saturated bromine water, diluting the mixture to 50 c.c., and filtering it through kieselguhr. The filtrate is then tested as described. Red and white wines examined all yielded a reaction for citric acid, although the quantity present was usually very small.TJsing a larger quantity of the wine (100FOOD AND DRUGS ANALYSIS 465 c.c.) the pentabromoacetone formed may be collected, dried over sulphuric acid, then washed with very dilute sulphuric acid, again dried, and weighed. For the detection of citric acid in milk, the latter is warmed with dilute sulphuric acid, the serum treated with tannin to remove soluble proteins, filtered, and the test applied to the filtrate. Fresh milk appears to contain more citric acid than does old and curdled milk. In the case of jams, a portion of the sample is diluted with water, acidified with a few drops of sulphuric acid, the solution is mixed with about twice its volume of alcohol, and filtered.The filtrate is neutralised with ammonia, evaporated to expel alcohol, again filtered after the addition of dilute sulphuric acid, and the filtrate used for the test.If the filtrate gives a precipitate with bromine, it is treated i s described under wine. Fruit juices also require a similar treatment before being tested. Colorimetric Determination of Pentoses and Methylpentoses in Wine. F. Schaffer. (Jahresversamml. Schweiz.Vereiizs. anal. Chem., Zurich, June 5, 1915 ; through J. SOC. Chem. Ind., 1915, 34, 1025.)-The following method is much more expeditious than that of Tollens, but requires strict adherence to the procedure described. Wines containing more than 3 per cent. of sugar must first be fermented. One hundred C.C. of the sample are evaporated to one-third of its volume, made up to 50 c.c., and heated under a reflux condenser for two hours with 22 C.C.of hydrochloric acid of sp. gr. 1.19. The liquid is next diluted to 100 c.c., and 92 to 93 C.C. are rapidly distilled (cork connections should not be used) and made up to 100 C.C. Five C.C. is transferred to a wide test-tube containing at least 0.2 grm. of resorcinol, 10 C.C. of hydrochloric acid of ap. gr. 1.19 are added, and after a rubber stopper, carrying a vertical tube of 6 mm.diameter, has been inserted, the test-tube is heated for ten minutes in boiling water. The liquid is then quickly cooled and compared with a standard solution which may be prepared by operating as above with a solution of arabinose of known titre, or with a wine containing a known amount of pentoses and methyl-pentoses, or a solution of (‘ Paper Red E ” (SOC.Chim. Ind., Basle) may be employed (0.01 per cent. solution diluted 7.5 times before use; this corresponds in colour to a 0.01 per cent. solution of pentose). The apparent pentose-content thus found must be corrected as described below for the influence of methylpentoses (rhamnose). To determine the latter, 5 C.C. of distillate (see above) is transferred to the test-tube, which contains sufficient pure vanillin to cover the tip of a knife-blade ; 10 C.C.of hydrochloric acid are added, and the tube is heated as before, but only for five minutes. The liquid is quickly cooled without agitation, and compared with a standard which may consist of a 0.01 per cent. solution of Merck’s indigocarmin. sicc. puZv. diluted 5.5 times before use, and treated, if necessary, with a trace of Chinese ink.The content of methylpentoses of the wine so found, multiplied by 0.6, must be subtracted from the apparent pentose content obtained by using resorcinol, to arrive at the true pentoee-con tent. In 200 natural wines (Naturwezne) examined, the ratio arabinose : rhamnose was never less than 2 : 1, whilst in ( ( marc ” wines, yeast wines, and raisin wines, it frequently ranged between 0.5 : 1 and 1 : 1.Alkaloids of Tobacco Extract. E. Noga. (Fachlich Mitt. osterr. Tabakregie, 1914; through J. Chem. SOC., 1915, 108, i., 711.)-Residues from Turkish tobacco466 ABSTRACTS OF CHEMICAL PAPERS were extracted systematically with water, and the extract evaporated in a vacuum. By the use of benzene, a small quantity of alkaloid, which was not volatile with steam, was obtained from this extract, and was separated by distillation into four fractions : (1) Nicotoine, C,H,,N, colourless, mobile liquid, with a peculiar intense odour resembling that of pyridine; boiling-point 208' C.; sp. gr. at 21'/4O C. 0.9545, ny = 1.5105. It gives the usual alkaloidal reactions.With hydrochloric, sulphuric, and picric acids, mercuric and platinic chlorides, it yields, in part, well-crystallised salts of definite melting-point. (2) Nicoteine. (3) Isonicoteine, Cl0HI2N2, colour- less, viscous, oily liquid, with a rather powerful, very persistent odour. I t is optically inactive, and has boiling-point 293" C. (darkening), sp. gr. a t 20"/4" C. 1.0984, & = 15749.With hydrochloric, sulphuric, and picric acids, mercuric, platinic, and auric chlorides, it yields, in part, well-crystallised salts. It forms a methiodide. I t is converted by oxidation into nicotinic acid, gives the pine-shaving reaction, and immediately decolorises potassium permanganate. It probably has the following constitution : CH--CH, CHQCR*CH \ c.c B I \N=CH/ \N ( CHJ .C H ~ (4) Nicotelline, CIOHsN2, melting-point 148" C., boiling-point above 300" C. (Zoc. cit.). Method for the Determination of Tobacco Resin and its Application to Various Tobaccos. J. von Degrazia. (Fachliclze Mitt. osterr. Tabakrcgie, 1914, 1-4; through J . Chem. Soc., 1915, 108, ii., 603-604.)-Two portions of 60 grms. each of the dry powdered tobacco are extracted in a Soxhlet apparatus with ether, about 100 grms.of ether being used for each portion. One of the ethereal solutions is then washed with water containing a small quantity of acid to remove nicotine, the ether evaporated, the residue dissolved in 90 per cent. alcohol, and the solution strongly cooled. After the separated wax has been removed, the alcoholic solution is boiled with the addition of 150 C.C.of 2 per cent. potassium hydroxide solution, and the mixture is extracted with ether ; the ethereal solution contains the tobacco resin, together with small quantities of ethereal oils. The solution usually has a reddish-brown colour, but in the case of Cserbel tobacco it has a green colour. After the ether has been evaporated, the residue of tobacco resin is weighed.The alkaline solution contains the resin acids and the resin alcohol ; the latter is in suspension, and is separated and weighed, whilst the resin acids are separated by extraction with ether in the usual way, the ethereal solution evaporated, the residue weighed, the acids dissolved in alcohol, and the tobaccic acid precipitated by the addition of alcoholic lead acetate solution.The lead salt is separated by filtration, and the filtrate mixed with ten times its volume of water acidified with nitric acid ; the B-tobacco acid which separates in flecks is collected, dissolved in ether, the solution evaporated, and the residue weighed. The y-tobacco acid is found by difference. The other portion of the ethereal solution is evaporated, and the residue extracted with alcohol to dissolve the u-tobacco acid; the alcoholic solution, which contains tannin and nicotine in addition to the acid, is evaporated,BACTERIOLOGICAL, PHYSIOLOGICAL, ETC. 467 the residue heated with very dilute potassium hydroxide solution, cooled, filtered, the filtrate neutralised, and the separated a-acid collected. The following table shows the composition of the reains from various tobaccos expressed as percentages of the latter ; the figures indicate that it is possible to differentiate between different kinds of tobacco : Wax. -- 0.59 9.55 0.55 0.77 1.32 0.69 1-48 1.00 Kind of Tobacco. Resin. 2.64 1-97 1-37 5.75 3-41 3.46 5.88 3-80 Havana ... ... ... ... Brazilian . . . ... ... ... Java ... ... ... ... Virginian . . . ... ... ... Hungarian . . . ... 1.. Cserbel (Nicotiana rustica) ... Dalmatian . . . ... ... Turkish ... ... ... . .. Total Besin. 4-65 3.36 4.45 7.59 5-15 4.57 7.86 7.00 Acids. Resin Alcohol. -_I__ -- 1.44 0.93 0.96 0.86 0.83 1.04 1-22 2.02 0.18 0.12 0.27 0.43 0.24 0.03 0-26 0.32 I 9.38 0.32 1 -75 0.44 0.65 0.04 0.45 0.81 0.01 0.02 0.10 0.02 0.02 None 0.05 0.07 (Cf. ANALYST, 1914, 39, 220.)