THE ANALYST. 41 9 ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOODS AND DRUGS ANALYSIS. Composition of Bananas. Niederstadt. (Chem. Zeit., 1908, 32, 936.)- The results of analyses of bananas imported into Germany are given in the following table : Fresh Green Bananas (as Imported). Ripe Yellow Bananas (from Jamaica). Water . . . . . . ... ... ... Fat.. . . . Nitrogenous matter . . . ... ... Sucrose ... ... Dextrose, etc. . . ... Starch ... . . - ... . . . Crude fibre ... . . . Free malic acid . . . ... ... ... Mineral matter ... ... ... ... . ... ... ... . . . . . . . . . ... . . . ... ... ... Per Cent. 72-16 0.11 1.35 11.68 7-26 5-22 0.94 0.27 1.01 Per Cent. 71.92 0-17 1.65 10.49 7.26 6-14 1.20 0.25 0.92 w. P. s. Determination of the Quality of Flours as regards their Suitability for Bread-making.M. Renner. (Zeits. Untersuch. Nahr. Genussm., 1908, 16, 234-238.)-The method desoribed by Liebermann (ANALYST, 1902, 27, 155), in which the quality of a flour is ascertained by measuring the volume to which the gluten expands when heated, was found to be capable of giving useful information as to the baking quality of a flour. It was noted, during the investigation here recorded, that very white, finely ground flours yielded gluten having a lower expansion value than the coarser ground flours. This value also diminishes the longer the flour is kept, and, if flour turns ‘‘ sour,” the decrease is very marked. w. P. s. Method of Estimating the Volume of Loaves, Pastry, etc. M. P. Neumann and P. Salecker. (Zeits. Untersuch. Nahr. Genussm., 1908, 16, 285-290.) --An apparatus is described by means of which the volume of a loaf of bread may420 THE ANALYST.be ascertained, the estimation being of some importance in comparing the baking qualities of various kinds of flours. It consists of a cylindrical glass vessel of sufficient capacity to hold the loaf under examination; the vessel is provided with a flange at the top, on which rests a glass cover, and this cover is secured by means of screw-clamps. A graduated burette holding 500 C.C. rises from the centre of the cover, and a tubulure in the side of the vessel, at about one-half the height of the latter, is connected by a length of flexible tubing to a pear-shaped reservoir which is provided with a tap. The whole apparatus is filled with water up to the zero mark of the burette (just above the cover of the vessel); the tap of the reservoir is then closed, and any excess of water remaining in the latter is poured away.By lowering the reservoir and opening the tap, 500 C.C. of water are withdrawn from the vessel into the reservoir, or, if a larger-sized loaf has to be measured, a larger quantity of water is removed; The cover is then unfastened, the loaf is placed in the vessel, and, after reclamping the cover, the water is allowed to flow into the vessel and burette. A simple calculation then gives the volume of the loaf. The latter should be coated previously with collodion or molten paraffin ; if the latter be used, a correction may be made for its volume, but this is scarcely necessary. A method is also described in which rape-seeds are used instead of water for determining the space occupied by the loaf, but the results obtained are not trust- worthy unless the greatest care be taken.A measure is filled with the rape-seeds, and the surface of the latter is levelled by passing a ( ( straight-edge ” over the top of the measure. The seeds are then transhrred to another vessel, the loaf is placed in the measure, which is next filled with the seeds and levelled as before. The volume of the excess of seeds is then measured, and gives the volume of the loaf. W. P. S. Detection of Benzoic Acid in Butter. G. Halphen. (Joum. Pharm. Chim., 1908, 28,201-203.)-The test proposed depends on the conversion of the henzoic acid into the ammonium salt of diamido-benzoic acid, which, in alkaline solution, has a brown-red colour.A portion of the sample of butter is melted together with sufficient lime-water to render the aqueous portion distinctly alkaline. After cooling, the aqueous layer is separated, rendered acid with phosphoric acid, and shaken out with ether, The ethereal extract is allowed to evaporate spontaneously, and the residue dried at the ordinary temperature. Two C.C. of concentrated sulphuric acid are then added, and the mixture is heated gently until the residue has dissolved; after adding 0.2 C.C. of fuming nitric acid, the solution is transferred to a dry test-tube and heated carefully over a small flame until sulphuric acid fumes appear in the tube. When cold, the mixture is diluted with 5 or 6 C.C. of water, which causes nitrous fumes to be given off; after again being cooled, the solution is treated with a saturated solution of sodium sulphite, which is added drop by drop until all the yellow vapours have disappeared.Ammonia is next allowed to flow over the surface of the solution, when an orange-red coloration is produced if benzoic acid is present in the sample, the intensity of the coloration being proportional to the quantity of benzoic acid. Unless the above procedure be adhered to, the test may fail, and in cases where a coloration is not obtained it is advisable to add a drop of ammonium sulphide to the ammoniacal eat-solution. If benzoic acid be present, a red coloration develops at the point of contact of the two liquids. w. P. s.THE ANALYST. 421 Detection of Synthetic Colours in Butter. R.W. Cornelison. (Journ. Arner. Chem. SOC., 1908, 30, 1478-1481.)-The author’s method consists in shaking thoroughly 10 grams of the melted fat with 10 to 20 grams of glacial acetic acid at about 35” C., when separation quickly takes place, and the acid layer can be drawn off, its colour noted, and portions treated with nitric acid, sulphuric acid, and other reagents. Nitric acid appears to give the best differentiation between the synthetic colouring matters and vegetable colours. The colour reaction between curcumin and sulphuric acid should be borne in mind. The following results were obtained by this method with various colouring matters, which were incorporated with a specially prepared pure butter to the extent of 1 part of dye in 100,000 of the butter : Dye. (Pure natural Soudan I.(pure) butter) Butter yellow (impure) Cerasine orange G. (Casella) Yellow 0. B. (Heller and Merz) (Heller and Merz) Annatto Yellow A. B. Curcumin Carrot ‘‘ Alderney butter colour ” (Heller and Merz) Ranson’s butter colour ( ‘‘ vegetable ”) Dandelion brand ” butter colour (‘‘ vegetable ”) Colonr of Acid Extract. Water- white Decided pink Very faint pink Strong greenish yellow Decided bright yellow Slight warm ochre-yellow Dull yellow Intense greenish- yellow Very faint greenish- yellow Brownish- yellow Yellow Yellow Concentrated Nitric Acid. Water-white Strong pink Faint pink Acid, yellow ; oil-globule, salmon-pink Acid, faint pink : oil-globule, salmon-pin k colourless Pink; fat Little change Dull ochre- yellow Faint yellow Strong pink Almost decolorised Almost decolorised Concentrated Sulphuric Acid.Faint pink on standing Strong clear pink Faint pink As with HNO, As with HNO, Brownish-pink ; oil faint pink Faint pink on standing Strong pink Faint pink on standing Strong pink As with HNO, As with HNO, Sulphuric Acid, and Ether to Clear Solution. Water- w hit e Pink Faint colour Brownish- yellow Pink Pink Very faint yellow Yellow Very faint yellowish - - -- A. R. T.433 THE ANALYST, Egg-Cognac. G. Heuser. (Zeits. Untersuch. Nahr. Genussm., 1908, 16, 290-292.)-1n the following table are given the results of analyses of four egg-liqueurs sold in Germany, the figures expressing grams per 100 C.C. : ... ... ... . . . Alcohol . , . Total solids ... .. . . . . Fat (ether extract) ... ... . . . Nitrogenous substances ... Total phosphoric acid ... Lecithin-phosphoric acid ... ... Ash ... ... ... . . . . . I ... Colouring matter ... . . . Boric acid ... ... . . . ... . . . . . . ... . . . Polarisation of 10 per . before inversion ... cent. solution ... { after inversion ... 23.00 41.00 4-93 0.29 0.033 0-022 0.078 Saffron 0 4.6' - 14O 11. Cognac. Egg- 15.11 35.56 4.43 2.34 0.144 0.085 0.400 Saffron 0.205 2-8O - 2.8' 111. Egg- Punch. 19.75 46.00 1.01 0.68 0.029 0.018 0.082 ? 0 3.5 O - 3.8" IV. ovos- Cream. 18.56 45.60 8.33 4-42 0.306 0-171 0.523 0 0 4.1" - 0.8" The ether extract contained, besides fat, flavouring substances such as spices, and in the case of No. I. some cream was present, Calculated from the amount of lecithin-phosphoric acid present (see ANALYST, 1905, 30, 245), the number of egg- yolks used in the preparation of each litre of these liqueurs was approximately: No.I., 1.5 to 2 ; No, II., 7 to 8 ; No. III., 1.5 ; and No. IV., 12 to 15. w. P. s. Detection and Estimation of Hexamethglenetetramine in Medicines. W. A. Puckner and W. S. Hilpert. (Journ. Anzer. Chenz. SOC., 1908, 30, 1471- 1474.)-The basic body hexamethylenetetramine, N = (CH, - N : CH,),, has of recent years come into use as an internal antiseptic and diuretic, and it possesses the property of decomposing, under suitable conditions, into formaldehyde and ammonia. Solutions of hexamethylenetetramine give with bromine-water an orange-coloured precipitate of the insoluble tetrabromide, which, when dried over caustic potash, gradually becomes converted to the canary-yellow dibromide, melting a t 198" to 200" C., with decom- position.Mercuric chloride forms with hexamethylenetetramine a white crystalline compound only slightly soluble in water, and which is decomposed by boiling with dilute acid, with formation of ammonia and formaldehyde. Hexamethylenetetramine is extracted from its aqueous solutions by chloroform, from which solution it separates in a crystalline form on evaporation. These tests serve for the detection of hexa- methylenetetramine when in admixture with other medicaments. For its determina- tion, a suitable quantity of the medicine (about 5 grams of a liquid preparation) is diluted with water to about 500 c.c., 10 grams of caustic potash are added, and the liquid boiled for half an hour to remove ammonia.The liquid is then rendered decidedly acid with sulphuric acid, and boiled for at least one and a half hours, inTHE ANALYST. 423 which time the hexamethylenetetramine is completely hydrolysed. The solution is next made alkaline by the addition of caustic soda, and the ammonia,, formed by the decomposition of the hexamethylenetetramine, distilled and titrated with decinormal acid, each C.C. of which represents 0.003479 gram of hexamethylenetetramine. A. R. T. The Estimation of Hydrastine. W. A. Puckner. (Chem. ZentrbZ., 1908, 2, 266-267.)-The following method of estimating hydrststine in the fluid extract and rhizome of hydrastis is recommended as giving better results than that of the United States Pharmacopceia : FZuid Extract.-Five C.C.are mixed with 5 C.C. of potassium iodide solution (20 per cent.) and 25 C.C. ot water, the mixture thoroughly shaken and filtered, and the precipitate washed twice in the beaker, and subsequently twice upon the filter, with portions of 5 C.C. each of a mixture of 1 C.C. of the potassium iodide solution and 19 C.C. of water. The united filtrate and washings are then shaken with three successive portions of 20 C.C. of ether, and once with 5 C.C. of ether, the united ethereal extraots filtered through cotton-wool and evaporated, and the residue dried at 95" to 98" C., and weighed. Rhizome.-Five grams of the powdered sample are mixed with 50 C.C. of ether, and the mixture treated, after ten minutes, with 2 C.C. of ammonia solution, and shaken at intervals for thirty minutes, The liquid is then filtered, the residue again extracted with 50 C.C.of ether, and the extract added to the first one and shaken successively with-(1) 2 C.C. of dilute hydrochloric acid* and 18 C.C. of water ; (2) 5 drops of dilute hydrochloric acid* and 10 C.C. of water ; and (3) 10 C.C. of water, The aqueous solution is now treated with ammonia, until alkaline to litmus, and extracted with three portions of 20 C.C. each of ether, the combined extracts evaporated at the ordinary temperature, and the residue dried at 98" to 100" C. C. A. M. Characteristics of the Fatty Oils from the Seeds of Certain Fruits. J. Kochs. (Chem. Bev. Fett- u. Harz-Id., 1908, 15, 256-257.)-The oils examined were obtained by extraction of the dried and crushed seeds with ether.The air-dried seeds of the red currant (Ribes rubrum) contained 16.9 per cent. of a yellowish-brown oil, which dried in seven hours at 50" C., forming a, hard almost colourless elastic film. The seeds of the berries commonly known as ( I hips" contained 9.63 per cent. of 8 thin drying oil of a yellow to orange-yellow colour. Tomato-seeds yielded 17.3 per cent, of a thin brownish-red oil, which had slight drying properties, and might be used as a food-oil. Raspberry-seeds contained 13.5 per cent. of a greenish- yellow oil, which had good drying properties, and might be used as a, substitute for linseed-oil. Strawberry-seeds contained 16.35 per cent. of a thick yellow oil, which had good drying properties; its high Reichert-Neissl value is noteworthy, but its low iodine value is to be attributed to oxidation of the oil brought about by the heat in crushing the seeds.Euonymus europsus yielded a reddish-brown, fairly thin, non-drying oil, with a, characteristic sharp odour. As the seeds yielded 35.2 per cent. of oil, the latter could probably be utilised with advantage in the manufacture of soap. The following table gives the analytical values of these oils : * Strength of acid not stated.424 THE ANALYST, Values. Specific gravity a t 15" C . . . Sa onificatioii value .. IoIine value ... .. Reichert-Meissl value . . Butyro-refractometer read - Solidification point, "C ,. ing a t 40" C. Fatty Acids .- Melting-poiiit, "C. . . Iodine value ... .. Mean molecular weight..Unaaponifiable matter, per cent. Currant-Seed Oil. 0'9120 171.3 1525 0.77 62" - 17.5 (semi-solid) 20 1695 266.2 2.31 Hip Oil. 0.9161 172.8 152-8 0.44 67" - 17.5 (viscous) 31 174.3 276.8 2 62 Tomato-Seed Oil. ~ 0-9200 183 '6 117'8 0.22 63" -9 (viscous) - 12 (semi-solid) 26-29 129-6 281.5 2.68 Rsspberry- Seed Oil. 0-9151 180'3 162.2 0.11 65" - 17.5 (semi-solid) 30 170.3 281.5 1.86 Strawberry- Seed Oil. 0,974 184.6 72.8 13'12 82" - 21 66 -3 241 2 -42 Oil of Ezcony- mus euiopams. 0,939 230'1 - 35.31 52" - 10 3s (1)egiiining) 70 105.3 251 2 (complcte) 5 *3:3 Characteristics of Japanese Tea Oil. M. Tsujimoto. C. A. M. (Chem. Rev. Fett- u. Harz-Ind., 1908, 15, 224.)-Two specimens of the seeds from the Japanese tea-plant (Thea sinensis) yielded respectively 23.90 and 26-23 per cent.of an orange-coloured oil with a somewhat unpleasant odour and bitter taste. The oil expressed from Tokyo seed gave the following analytical values : Specific gravity at 15" C., 0.9178 ; acid value, 0.74 ; saponification value, 191.91 ; iodine value (Hubl), 9042 ; Hehner value, 95.6 ; Reichert-Meissl value, 0.66 ; and refractometer reading at 20" C., 1.4707. Fatty Acids.-Specific gravity a t 98' C., 0.8445 ; melting-point, 33.5" C. ; neutralisation value, 197.60 ; mean molecular weight, 283.91 ; and iodine value, 92-86. The oil was readily soluble in the ordinary solvents for fats. I t solidified at - 10" C. In the elaidin test it became viscid after two hours, and solidified to a butter-like mass after twenty-four hours. It could be used for the same purposes as tsubaki and sasmqua, oils (ANALYST, 1908, 238), which it closely resemble sin many respects; but it would require careful refining to remove saponine-like substances before being suitable for food.So far it has not become a coinrnercial product. C. A. &I. Separation and Determination of Salicylic Acid and Methyl Salicylate in Foods, ete. H. D. Gibbs. (Jozmz. Amer. Chem. Soc., 1908, 30, 1465-1470.)- Methyl salicylate is added as a flavouring agent to various beverages, foods, and medicinal preparations, the synthetic product and the oils of gaultheria and sweet birch being employed for this purpose. These frequently contain a small quantity of free salicylic acid, the author having found quantities up to 0.025 per cent., while the free acid itself may also be simultaneously added to the food, and so a method for the separation of the acid and ester is desirable.The problem is complicated by the fact that methyl salicylate hydrolyses to a varying extent when added to certain substances, particularly in presence of alkalies and in a warm atmosphere. TheTHE ANALYST. 425 author has devised the following methods for the examination of the ester, and for the determination of salicylic acid and its methyl ester when present together in foods, etc. : Dctcriitination of Salicylic Acid in Methyl Salicy1ate.-This is best carried out by direct titration with standard sodium bicarbonate solution, using congo-red as the indicator. Five to twenty cc. of the ester are shaken with an equal volume of neutral distilled water in a stoppered flask, and sodium bicarbonate (free from normal carbonate) run in until the end-point of the titration with congo-red is permanent on shaking.This method gives results agreeing closely with those obtained by thoroughly extracting 10 C.C. of the natural oil or synthetic ester with Tc sodium bicarbonate solution, removing any traces of ester from the alkaline extract by shaking three times with chloroform, and then acidifying the alkaline solution, extracting the salicylic acid with chloroform, evaporating off the solvent, and determining the salicylic acid in the residue colorimetrically, as described in the 77. S. Dcpt. Agric. Bull. 107 (1907), p. 180. Sepa,ration and Estimatio.12 of Salicylic Acid and Methyl Salicylate i i z Foods, Beverages, etc.-The substance is made strongly alkaline to congo-red with approxi- mately normal sodium bicarbonate, and if the mixture is not homogeneous the aqueous solution is separated, and the extraction with alkali repeated until complete.All the salicylic acid and small amounts of methyl salicylate are now in the alkaline solution, which is shaken three times with chloroform to remove the methyl salicylate, and then acidified with diluted sulphuric acid, the salicylic acid removed by solution in chloroform, and estimated colorimetrically. The extraction of the material with the alkaline liquid should be carried out at a temperature not much above 30" C., and preferably lower, to avoid hydrolysis of the methyl salicylate. The methyl salicylate present in the material is extracted from it, after removing salicylic acid as above, by shaking several times with chloroform, and to the mixed chloroform extracts the small quantity of methyl salicylate, already extracted in the first part of the process, is added.The chloroform is next heated for half an hour under a reflux condenser with excess of 25 per cent. caustic soda solution to saponify the ester, the chloroform removed by evaporation, the residual solution then suitably diluted, and the colorimetric estimation of salicylic acid carried out on an aliquot portion. The above method has been successfully applied to the examination of such preparations as cod-liver oil emulsion, in which case the alkaline solution is separated from the emulsion by rapid centrifugalisation of the mixture.After thus removing the free salicylic acid, the methyl salicylate in these and certain other .mixtures is best separated by steam-distillation of the material acidified with sulphuric acid. Since a partial hydrolysis of the ester takes place in this acid solution, the steam- distillation must be continued until no more salicylic acid passes over into the distillate. The author shows that methyl salicylate undergoes rapid hydrolysis in presence of sodium hydroxide, and that this decomposition is also considerable in presence of sodium carbonate, though sodium bicarbonate free from the normal carbonate is practically without action. The hydrolysis in all cases increases greatly with theTHE ANALYST. temperature. For the above reasons it is necessary to employ sodium bicarbonate solutions for the extractions in carrying out the foregoing methods, and to keep the solutions as cool as practicable.A. R. T. Malt Analysis : Determination of Extract.-11. A. Hunicke. ( J o ~ m . Amer. Ghern. SOC., 1908, 30, 1431-1443.)-1n diluting and filtering a malt mash for the determination of the density of the wort, phenomena of mechanical absorption and colloidal adsorption come into play, particularly with the coarser forms of grist, so that the various fractions of the filtrate contain different amounts of dissolved matters. When, in the ordinary way, the cooled mash is diluted to 450 grains, stirred thoroughly and filtered, the concentration of the first, fraction of 75 C.C. is well below the average concentration.If the mash be stirred and allowed to stand for twenty-four hours before filtration, the extract is considerably higher than if it be filtered a short time after making; but if, after twenty-four hours, a quantity of the wort be pipetted off without filtering, it contains several per cent. less solids than the filtered wort, showing an adsorption of the solute by the fibrous colloids of the mash. In the case of coarse grists, the extraction of the viscous dextrins from the residual grains is very slow, the difficulty being chiefly a mechanical one. Of course, the very object in working a coarse griat in the laboratory is to endeavour to estimate this retardation, the coarse grist analysis being a technical and not an absolute deter- mination. The author describes the procedure he adopts in malt analysis in the following terms : Weigh off 50 grams of malt, and grind either to finest powder or in a, Seck mill at 25" C., according to requirements, avoiding all loss in the process.Add f o the malt in a copper beaker 200 C.C. of water at such a temperature that the mixture shows 45" C. Place in water-bath, stirring uniformly at regular intervals, preferably by a rotary motion of the thermometer, six rotations every five minutes. After thirty minutes at 45" C., the temperature is raised 1' C. per minute until 70" C. is reached, and the whole is kept at 70°C. for one hour. Ten C.C. of mercuric chloride of such concentration that the whole wort will contain 0.002 per cent. are added, and 200 grams of water at 70" C. The whole is stirred during the addition of the water (ten minutes), and for five minutes after.It is then cooled and made up to 450 grams. After stirring vigorously, the wort is filtered off, the first 150 C.C. are returned to the filter, and 300 C.C. are taken for the determination of the density. This method gives good results, provided the itirring and the heating up of the mash be carefully controlled. The dilution of the mash, at a temperature of 70" C., instead of after cooling, facilitates the diffusion of the dissolved matters, and gives rather higher results in consequence. By this method of mashing, the first fraction of 75 C.C. is of higher concentration than the second, the subsequent fractions increasing as in the ordinary method. This is attributed to the fact that the first fraction filters so rapidly that adsorption by the fibres of the paper has not time to take effect. J.F. B. Preliminary Tests for Maple Products. A. P. Sy. (Jouriz. Amer. Chent. Soc., 1908, 30, 1429-1431.)-Three new tests are described for the preliminary examination of maple products : (1) Golour.-Maple products contain besidesTHE ANALYST, 427 caramel, which ia insoluble in amyl alcohol, other colouring matters which are soluble in that medium and also in water. The test is carried out in the following manner: 15 C.C. of syrup, or 15 grams of sugar and sufficient water to make 15 c.c., are placed in a test-tube, 3 C.C. of pure amyl alcohol are added, and 1 C.C. of a 20 per cent. solution of phosphoric acid; the latter reagent decreases the solubility of the coloured substances in the water.Pure maple products colour the amyl alcohol layer a decided brown ; adulterated samples give fainter colours according to the quantity of maple present ; cane-sugar products, coloured with caramel, do not react. (2) Foam Test.-When heated and ignited for the ash determination, maple products form a considerable amount of foam, whereas cane products do not. The foam test is carried out by placing 5 C.C. of a syrup in a narrow cylinder graduated in 0.1 c.c., 10 C.C. of water are added, and the mixture is shaken vigorously for half a minute, allowed to stand for ten minutes, and the volume of the foam is read off. Maple syrups give from 3 to 6 C.C. of foam, average of sixty tests, was 4-1 c.c., although the two abnormal samples with a peculiar odour gave much less. (3) Volzme of Led Bnsic Acetate Precipitate.-The following procedure gives fairly concordant results : 5 C.C.of syrup, or 5 grams of sugar made up to 5 c.c., are placed in a 25 C.C. measuring cylinder ; 10 C.C. of water and 2 C.C. of basic lead acetate are added ; the liquid is mixed and then allowed to stand for twenty-four hours. For pure maple products the volume of the precipitate at the end of this time should be over 3 c.c., and is generally over 5 C.C. ; the average of twenty-three pure samples was 8-5 C.C. for syrups and 7.9 C.C. for sugars. J. F. B. Estimation of Dry Substance by the Refractorneter in Liquid Saccharine Food Products. H. Bryan. (Jown. Anter. Chern. SOC., 1908, 30, 1443-1451.)-The Abbe high-temperature prism refractometer has come into use for the estimation of dry substance in sugar-house products.Main (Intern. Sugar Jozwn., 1907,9,481) and Geerligs (ibid., 1908, 10, 68) have prepared tables showing the relations between the dissolved solids and the refractive indices, and the latter gives the corrections to be applied for temperature. The chief source of error is in the non-sugar constituents and the saline matters, so that the results with molasses are not so accurate as with purer products, but still they are quite available. The author has carried out a large number of determinations of dry substance, by direct drying on sand and by the refractometer, in samples of maple and cane syrups, honeys, cane and beet molasses, and glucose syrups, using Geerligs’ tables and the temperature corrections.He finds that in most cases the refractometer results are slightly lower than the results obtained by the direct method, but only in the case of honeys are the differences at all considerable. The refractometer results agree more closely with the direct results than the values obtained by the specific gravity method from Brix’s tables, which are always too high. The method of direct drying is so inconvenient, and the results of duplicate determinations are so variable, that there can be no question as to the great utility and substantial accuracy of the refractometer method. J. F. B.428 THE ANALYST. Determination of Sugar in Meats. A. Lowenstein and W. P. Dunne. (Jozmz. Amer. ClLeiu. Soc., 1908, 30, 1461-1465.)-The authors describe a method for the estimation of sugar in meat, which is simpler and more accurate than the official American method, and which allows of the estimation of the nitrates in the same extract.Three hundred grams of finely-divided meat are placed in an enamelled iron casserole of about 1 litre capacity, 200 to 300 C.C. of water are added, and boiled for thirty minutes. The mass is cooled, the liquid is strained through muslin, and the residue is squeezed and boiled up again with 200 C.C. of water for fifteen minutes. The second extract is added to the first and evaporated in an iron casserole over a free flame. When saltpetre is to be determined, the extraction of the meat is repeated until the liquid fails to respond to the diphenylamine test; two extractions are generally sufficient.The meat and cloth are thoroughly washed with hot water, and the united extract is concentrated to 300 to 400 C.C. The liquid is transferred to a separating funnel, and the fat is separated, the aqueous solution being returned to the casserole. The fat is washed several times with 50 C.C. of hot water each time until the washings fail to respond to the diphenylamine test. Evaporation is continued down to about 150 C.C. The extract is then transferred to a 200 C.C. graduated flask. Aliquot portions may then be taken for the determination of reducing sugars before and after inversion and of the nitrates. The sugar is preferably determined volumetrically by Pavy’s solution. J. F. B. Detection and Estimation of Free Mineral Acids in Red Wines.C. Astre. (Bzdl. SOC, Clzim., 1908 [iv.], 3, 928.)-If a red wine be dialysed into distilled water, the colouring matter does not pass through the membrane for a long time (about one and a half hours), while at the end of that period the outside liquid contains a large proportion of the salts and free acids in the wine. By working with the same dialyser under the same conditions, it is possible both to detect and to estimate free mineral acids in wine. For this purpose, 80 C.C. of the wine are placed in the dialyser, which is immersed in a vessel containing 50 C.C. of distilled water at the same level as the liquid within the membrane. After one and a, half hours the acidity of the exterior liquid is estimated by adding to 10 C.C. of it 10 drops of glacial acetic acid, and titrating the mixture with ;G potassium hydroxide solution, with tropEolin 00 as indicator.Natural wines, even when containing an addition o€ tartaric acid, and plastered wines, only require 1 or 2 drops of the alkali solution, whereas wines containing a few decigrams of mineral acids require an appreciable- quantity of the reagent before becoming yellow. By making comparative tests with wines containing a known quantity of a mineral acid, approximately quantitative results may be obtained. After each test the dialyser is washed successively with a 10 per cent. solution of sodium bisulphite, with a 1 per cent. solution of potassium hydroxide, dilute acetic acid, and water. C. A. M. The Indirect Estimation of Benzoic Acid in Wine.G. Favrel. ( A m . dc Chivz. ilnul. AppZ., 1908, 13, 343-345.)-The method is based upon the fact that benzoic acid may be extracted by means of ether from its solution in dilute (10 to 12 per cent.) alcohol. If then a wine be treated with a known quantity of --$ sodiumTHE ANALYST 429 hydroxide solution in excess of that required to neutralise its acidity, and then with the quantity of benzoic acid exactly equivalent to the added alkali, the acidity of the solution of extracted benzoic acid should be the same as that of the original wine. A standardised solution of benzoic acid ,is prepared by treating a known weight of that acid with 25 C.C. of 90 per cent. alcohol, and titrating the liquid with & sodium hydroxide solution with phenolphthalein as indicator. In estimating the acidity of a wine 10 C.C. of the sample, previously freed from carbon dioxide, are treated with 20 C.C. of zc sodium hydroxide solution, and then with the quantity of benzoic acid exactly equivalent t o the alkali. The liquid is then shaken in a stoppered cylinder with two successive portions of 40 C.C. of neutral ether, and the extracts united and evaporated on a water-bath at 70" C. in such a way that the ether distils without ebullition. As soon as the ethereal solution has been reduced to a fourth of its original volume, the distillation is stopped and the evaporation completed at a low temperature in a basin 12 em. in diameter. The distillation flask is rinsed first with 10 C.C. and then with 5 C.C. of 90 per cent. alcohol, and the washings added to the liquid in the basin, the contents of which are finally titrated with ZG sodium hydroxide solution after the addition of two drops of phenolphthalein solution as indicator. The number of C.C. required multiplied by 0.00245 gives the acidity in the 10 C.C. of the original wine. The method gives results in close agreement with those obtained by direct titration of colourless wines, and will be of use in determining the acidity of dark-coloured wines. C. A. M.