48 THE ANALYST+ ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOODS AND DRUGS ANALYSIS. Detection and Estimation of Benzoic Acid in Ketchups, Fruits, and Ciders. H. S. Reed. (Journ. Amer. Clzenb. SOC., 1907, 29, 1626-1629.)-For the detection of benzoic acid a modification of Mohler's test is used. One hundred grams of the substance under examination are acidified with dilute sulphuric acid, and then thoroughly ex- tracted with chloro- form. The chloroform solution is placed in the reservoir A of the apparatus shown. This part of the apparatus is detached, and con- nected directly with a suction-pump, and the chloroform is allowed to evaporate spontane- ously under a current of air. When the evaporation is complete, the reservoir is placed in a desiccator until perfectly dry.It is then attached to the vessel B and submerged to the neck of the funnel C in a sand-bath, about 12 C.C. of N sodium hydroxide being placed in B. The suction-pump is connected to the latter vessel, and the sand-bath is heated to a temperature of 145' C., and then raised to 260° C. The benzoic acid is thus sublimed and retained in B as sodium benzoate. This solution is next trans- forred to a separating funnel, acidified with sulphuric acid, and extracted with chloroform. The chloroform solution of the benzoic acid is rendered alkaline by the addition of alcoholic potassium hydroxide solution, and the mixture is evaporated. The residue of potassium benzoate is heated with concentrated sulphuric acid until white fumes are given off, and the sulphobenzoic acid formed is converted by the use of potassium nitrate to m-dinitrobenzoic acid. The acid solution of the latter is cooled, diluted with water, again cooled, made alkaline with ammonia, and treated with hydrogen sulphide gas.A cherry-red coloration is produced by the reduced product, the ammonium salt of m-diaminobenzoic acid. The method proposed for the estimation of benzoic acid is based on two conditions : the first, that chloroform extracts benzoic acid from many other vege- table acids (malic, tartaric, oxalic, etc.) in aqueous solution; and the second, thatTHE ANALYST. 49 calcium benzoate is far more soluble in cold water than the calcium salts of the above- mentioned acids, The chloroform solution obtained from the sodium benzoate solution contained in the part B of the apparatus is allowed to evaporate spontaneously in a beaker.To the residue are added 25 C.C. of milk of lime containing 0.0145 gram of metallic calcium per C.C. This milk of lime is prepared by acting on water with rrietallic calcium. Another 25 C.C. of the milk of lime are placed in a second beaker, and subjected to the same treatment as the former. The contents of the beakers are evaporated to dryness, and the residues are taken up with 25 C.C. of water, filtered, and washed with 15 C.C. of water. The filtrates are collected in platinum basins, evaporated to dryness, and the residues ignited over a blast burner. The amount of calcium oxide in each basin is then obtained by titration in the usual way. The quantity present in the blank is subtracted from that found in the test, and the difference is calculated into benzoic acid.w. P. s. A New Constant for Butter Analysis and the Detection of Cocoanut Oil in Butter-Fat. T. R. Hodgson. (Chcm. News, 1907, 96, 273-274, 288, 297.)- The constant described in this paper depends on the oxidation with potassiuni permanganate of the acids obtained from butter-fat. The author finds that the quantity of oxygen required to osidise a given quantity of the saponified fat is invariable. The method used is as follows : One gram of the filtered fat is placed in a flask and saponified with 25 C.C. of alcoholic potassium hydroxide solution, the mixture being boiled for thirty minutes under a reflex condenser to ensure complete saponification. The solution is then evaporated to dryness on the rater-bath.Fifty C.C. of water are added to the residue, and the whole again evaporated to dryness. The residue is then dissolved in water and diluted to 1 litre. Twenty C.C. of the solution are transferred to a beaker, 50 C.C. of 2; potassium perinanganate solution and 50 C.C. of 50 per cent. sulphtlric acid are added; the beaker is heated in a water-bath for thirty minutes, and the excess of permanganate is then titrated back with ferrous ammonium sulphste solution. Butter-fat uses 41.8 C.C. of & permanganate solution. From this figure the number of grams of oxygen required for the oxidation of 1 gram of butter-fat can be calculated, and the result multiplied by 100 gives the " oxygen equivalent" of the fat. Twenty samples of cocoanut oil examined required from 18.0 to 26.8 C.C.of :;i permanganate, correspond- ing with '' oxygen equivalents " of from 72.0 to 107.2, the average value being 88-66. Results of analyses of test samples consisting of butter and cocoanut oil are given, from which it is seen that the method enablee the quantity of the latter oil present This operation is repeated until all the alcohol has been removed. This number for butter-fat is 167.2. t o be estimated with a considerable degree of accuracy. w. P. s. Some Algerian Olive Oils. L. 'Arehbutt. (Journ. SOC. Clzenz. I d . , 1907, 26, 1185-1186.)-The following results were obtained on the analysis of twelve authenti- cated*saniples of Algerian olive oil : Specific gravity, at 15.5" c., 0.9165 to 0.9178 ; free acid (as oleic), 0.4 to 4.5 per cent, ; saponification value, 18.90 to 19.13 ; iodine value (Wijs), 82.4 to 90.4; unsaponifiable matter, 0.72 to 0-98 per cent.No arachidic acid was found in the samples. When shaken with nitric acid (specific50 THE ANALYST, gravity 1.375), the oils gave no coloration immediately ; but, after standing for one hour, one oil turned greenish-brown, another developed a dark greenish, and the remainder pale brownish tints. The iodine values of some oils from named varieties of olives are also given. Whilst exceptionally high values are given by a few varieties grown in certain districts, the oils from other varieties yielded normal results. For example, the variety Chemlali-Sfax (Tunis) gave a value of 80.2 ; Chetui Bizerte (Tunis), 91.1 ; Chetui Mornag, 94.7 ; Rousette (Algeria), 90.1, etc.(Compare ANALYST, 1907, 32; 257.) Tvv. P. s. Niam Fat. J. Lewkowitsch. (Joz~rn. SOC. Chcm. Ind., 1907, 26, 1265-1266.) -The author has investigated the fruits of Lophira alatn from Sierra Leone. They yielded 61.5 per cent. of kernels and 38.5 per cent. of husks. The kernels, when extracted with ether, gave 0.8 per cent. of a black resinous body, which separated from the fat after evaporating off the ether. The fat freed from resin amounted to 41.19 per cent. of the kernels, or 19.18 per cent. o€ the whole seeds. I t is a soft buttery mass melting at 24' C., known in West Africa as Niam fat " or (' Meni oil." I t has an unpleasant taste, due probably to traces of dissolved resin, and is used by the natives for culinary purposes and as a hair-oil, Below are some of the characters of two samples : Specific gravity at 40" C.(water at 40' C. = 1). .. Saponification value ... ... ... ... Unsaponifiable matter ... ... ... ... Mean molecular weight of fatty acids ... ... Solidifying point of the fatty acids (titer test). .. Acid value ... ... ... ... ... ... Iodine value ... ... ... ... ... Fat extracted from Kernels. 0.9105 18.54 195.6 1.49 per cent. 68.4 - Fat prepared by Natives. 0.9063 5.78 190.1 1.38 per cent. 78.12 283.7 42.5" C. J. F. €3. Application to Solid Fats of Renard's Test for Peanut (Earthnut) Oil. W. B. Smith. (Jozwn. dnzer. C h m . SOC., 1907, 29, 1756-1757.)-Tolman's modifica- tion of this test consists in saponifying at least 20 grams of the sample with alcoholic potash, and treating the resulting soap with lead acetate, when the extraction of the lead soap with ether dissolves the lead oleate, linoleate, etc., and leaves the lead stearate, palmitate, and arachidate behind.On acidifying the insoluble portion with hydrochloric acid, the fatty acids are freed, and are separated and dried, dissolved in hot alcohol (90 per cent;.), and the solution cooled to 15" C. The arachidic acid crystallises out, is weighed, and its melting-point determined ( = 75" C. for the pure acid). The weight of arachidic acid, multiplied by 20, gives the approaimate amount of peanut oil present. I n using this method for the examination of solid fats (such as lard) containing large percentages of fatty acids, a precipitate is invariably obtained from alcohol,THE ANALYST.51 though it is frequently gelatinous and quite unlike the crystalline arachidic acid. Pure lard gave, by the above test, crystals melting at 54.3' C., while when mixed with 10 per cent. of peanut oil, the crystals obtained melted at 55" C., and the test thus failed to detect this proportion of peanut oil. When, however, the crystals first obtained were dissolved in alcohol and recrystallised, the melting-points of these products were 55' C. and 77" C. respectively, the latter evidently consisting chiefly of arachidic acid. I n the case of solid fats, therefore, the first crystals from alcohol should always be recrystallised until a constant me1 ting-point is obtained. A. R. T. Composition of Camembert Cheese. P. Buttenberg and F.Guth. (Zeit. Untemzich. Nahy. GenzLssm. , 1907, 14, 677-682.)-Analyses of twenty-two brands of Camembert cheese are recorded, froiii the results of which it is seen that this cheese is usually manufactured from full milk. I n fifteen of the samples the ratio of the fat to the fat-free dry solids varied from 1 : 0.86 to 1 : 1.76 ; in five samples (probably made from partially skimmed milk) the ratio was from 1 : 2 to I : 3 ; two other samples, obviously made from skimmed milk, contained 8.48 and 9.13 per cent. of fat respectively, the ratio varying from 1 : 1-32 to 1 : 1.41. I n only five of the samples was the amount of water below 50 per cent.; one of the above-mentioned skim-milk cheeses contained 61.6 per cent,. of water. w. P. s. The Composition of Extract of Crab.D. Ackermann and F. Kutscher. (Zed. I/'nterszcch. Nahr. Cfewss. , 1907, 14, 687-691.)-1n addition to the substances mentioned previously (ANALYST, 1907, 32, 256) as occurring in extract of crab, the authors have separated and identified the following bases : Crangitine (C,3H,oN,0,), crangonine ( C13H,,N,0,), neosine, and methylpyridine-ammonium hydroxide. The presence of the latter is worthy of note ; hitherto this base has been deemed peculiar to the vegetable kingdom. It has certainly been found in human urine (probably derived from vegetable foods containing pyridine bases), but never in meat extracts. w. P. s. The Application of Folin's Creatin and Creatinin Method to Meats and Meat Extracts. A. D. Emmett and H. S. Grindley. (JOWZ. Biol.Chem., 1907, 3, 491-516.)-1n reply to criticisms of Hehner (Phnrm. Jown., 1907, 78, 683) on Folin's method, the authors describe experiments in detail, from which they conclude that the method, when suitably modified, gives as accurate results with meat and meat extracts as with urine. They find that an increase in the amount of picric acid does not affect the estimation of the original creatinin, but that as a rule it does produce a difference when the converted creatin is also present. The quantity of picric acid solution (1.2 per cent.) should therefore be increased to 30 C.C. in estimating the dehydrated creatinin, but the 15 C.C. recommended previously (ANALYST, 1907, 32, 171) should be retained for the original creatinin estimation. Contrary to Hehner's experience, they find that variations, in the proportion of 10 per cent.alkali, make but; little difl'erence (though slightly lower results are obtained with 5 C.C. than with 10 or 15 c.c.) in the estimation of the original52 THE ANALYST, creatinin ; whilst in the case of the converted creatin 10 C.C. give better results than 5 c.c., but no better than those obtained with the large excess of 15 C.C. The method now used in the authors' laboratory has been modified as follows : The estimation of the ready-formed creatinin is made in the Bame way as described (Zoc. cit.). Aliquot portions of the solution are then made up to 10 C.C. (or evaporated if necessary), mixed with 10 C.C. of N-hydrochloric acid, and heated for thirty minutes in an autoclave at 117" to 119' C.They are next cooled and made up to 100 c.c., and aliquot portions transferred to 500 C.C. flasks, and treated with 30 C.C. of the picric acid solution and 10 C.C. of 10 per cent. sodium hydroxide solution, the colour produced being matched with $ bichromate solution, as before described. In experi- ments with solutions of pure creatin, the results of six colorimetric estimations ranged from 94.2 to 95.8 of the theoretical quantities. C. A. M. Characteristics of Segura Balsam. Utz. (Chem. Rev. Fett.-u. HCLYX. I?d, 1907, 14, 295-296.)-1t is stated that copaiba balsam is no longer being adulterated with gurjun balsam alone, but with segura balsam, or a, mixture of the latter with gurjun balsam. The sample of segura balsam examined was a dark-brown, sticky, viscous mass resembling Peru balsam. On warming it became very mobile, and emitted yellow vapours when strongly heated.I t was soluble in chloroform, benzene, petroleum spirit, and carbon tetrachloride, partially soluble in alcohol, and formed a dirty-green emulsion with ammonia solution. A dark red-brown coloration was produced on adding concentrated sulphuric acid to its solution in one of the above solvents, whilst with nitric acid (specific gravity 1.40) a yellowish-green coloration was obtained. Vanillin and hydrochloric acid gave a red-brown coloration, and stannous chloride a light-brown colour. I t had the following analytical values: Specific gravity at 15' C., 1,0337; acid value, 14.04; and saponification value, 92.66. The addition of segura balsam to gurjun balsam would raise the specific gravity and ester value, and lower the acid and saponification values.The values of Para copaiba balsam would be similarly affected, except that the saponification value (cold method) would be somewhat increased. The dis- tillation of segura balsam yielded about 5 per cent. of water, followed by 30 t o 40 per cent. of a light-brown ethereal oil of pleasant odour, and having the following constants : Specific gravity at 15" C., 0.9451 ; polarisation in 200-mm. tube, - 19.00 ; and refractive index, 1.4992. This oil behaved like the original balsam towards the above-mentioned reagents. This distilled oil is probably used in admixture with segura balsam, so as to prevent the latter affecting the values and the colour of copaiba balsam too much.I t had a pleasant aromatic odour. c. A. N. The Constituents of the Essential Oil of Nutmeg. F. B. Power and A. H. Salway. (Proc. Chem. Soc., 1907, 23, 285.)-The material employed in this investigation consisted of an oil which was distilled from unlimed Ceylon nutmegs of good quality, and was obtained in a yield of 6.94 per cent. I t had d 15"/15"=0*8690; a,, + 38'4' in a 1-decim. tube ; acid value, 0-81 ; ester value, 3.15. The more complete examination of the oxygenated constituents of the oil was accomplished by means of a product, designated heavy oil of nutmeg, from which the terpenes had been to aTHE ANALYST. 53 large extent removed. This had d 2O0/2Oo=1*102; a,+1°17’ in a I-decim. tube; saponification value, 6-10. Essential oil of nutmeg contains the following substances : (1) Eugenol and (2) isoeugenol (about 0.2 per cent.); (3) d-pinene and (4) d-camphene (about 80 per cent.) ; ( 5 ) dipentene (about 8 per cent.) ; (6) d-linalool, (7) d-borneol, (8) i-terpjneol, arid (9) geraniol (about 6 per cent.); (10) a new alcohol, yielding on oxidation a diketone, C,H,,O, (a very small amount); (11) a trace of an aldehyde resembling citral, but yielding a P-nap~~thacinchoninic acid derivative melting at 248” ; f12) safrole (about 0% per cent.); (13) rnyristicin, C,,H,,O, (about 4 per cent.); (14) myristic acid, in a free state (about 0.3 per cent,), and apparently a small amount in the form of esters; (15) formic, acetic, butyric, and octoic acids, and a new monocar- hoqjZic acid, C,3H,803, all in the form of esters, and in relatively small amount.This investigation has also shown that the portion of nutmeg oil which has hitherto been designated myristicol ” is it mixture of alcohols, of which terpineol appears to be the predominating constituent. Application of Lovibond’s Tintometer to the Analysis of Spirits, Etc. E. Westergaard. (Jozmt. Soc. Chem. Id., 1907, 26, 1226-1228.)-The author has employed Lovibond’s tintometer for the colorimetric estimation of the aldehydes and furfural in whisky, The usual standard solution of aldehyde was prepared and mixed at various dilutions with the magenta-sulphurous-acid reagent. The mixtures were placed in a, water-bath at 25” C. for fifteen minutes, and then examined in the $-inch cell in the tintometer. I t was found to be impossible to match the colours satisfac- torily, but good results were obtained by neutralising the colours with yellow and blue glasses until no colour was visible in the tintometer. I n the case of the magenta colours it was found that the quantity of yellow required was always double that of the blue. The results are calculated from the empirical equation y = 0-251 4x1*lG3, where y = the value of the yellow glasses, and x=cgm. of aldehyde per litre. In an exactly similar manner the colours produced by furfural and aniline acetate were estimated by neutralisation, but in this case the blue glasses required were in excess of the yellow. The equations were determined-for the blue y = 1.8138xU !‘2fi!‘4, and for the yellow y=1*5805zO*SS~, where x=mgm, of furfural per litre. The success obtained in thus recording numerically the colour reactions of aldehyde and furfural led the author to make preliminary investigations with two other colour-tests-viz., ammonia and Nessler’s solution and the nz-phenylenediamine reaction for nitrites. Both these colours could be successfully neutralised by blue and red glasses; the estimations were made in 1-inch cells, but a still deeper cell would be desirable. In performing all these colour-tests against numerical standards, the great influence of temperature and time on the intensity of the colours and the standard conditions adopted must be rigorously observed. J. F. B.