THE ANALYST. 357 ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOODS AND DRUGS ANALYSIS. The Unsaponifiable Constituents of Cocoanut Fat and its Detection in Admixture with Butter. H. Matthes and E. Ackermann. (Ber. Deut. Chem- Ges., 1908, 41,2000-2001.)-The authors have ascertained that cocoanut fat contains, in addition to the ordinary known phytosterol, a second phytosterol giving similar colour reactions, but combining by addition with two molecules of bromine, and yielding an acetate which forms a sparingly soluble tetrabromide which is charac- teristic. For instance, from 1 kilogram of cocoanut fat there were obtained 1-25 grams of crude phytosterol, and 0.25 gram of a liquid residue, as the result of a twice- repeated saponification, and shaking out with ether in a Hagemann's apparatus.The crude phytosterol melted at 135"-145" C.; after acetylation the product melted at 126"-128" C. The acetylated product was then brominated in presence of ether and glacial acetic acid, and the characteristic tetrabromide then separated out in the form of thin plates, melting with decomposition at 180"-183" C., and containing 39 per cent. of bromine. From the filtrate, after evaporation and recrystallisation from alcohol, crystalline aggregates of phytosteryl acetate dibromide, melting-point 132O-135" C., were obtained. Since butter contains only cholesterol, and as the latter does not form an acetate tetrabromide, it is suggested that this reaction may possibly be available for the identification of cocoanut fat, or at all events of vegetable fat, as am adulteration in butter, and the method would possess the advantage over the present indirect methods that an actual derivative of a constituent of the adulterating substance could be isolated.J. F. B. The Valuation of Indian Hemp. D. Hooper. (Plznrm Jozwn., 1908,81,80.)- The gradual deterioration of hemp drugs on keeping is to be attributed to the oxidation of the red-oil, cannabinol, and probably also of more volatile constituents. Hence a determination of the iodine value and percentage of unsaponifiable matter in an extract affords a criterion of the extent of deterioration. I n order to remove organic acids, resins, fixed oils, and lighter terpenes, 1 to 2 grams of the extract are saponified for two hours with 20 C.C. of alcoholic potassium hydroxide, the liquid then evaporated to dryness, the residue washed with water, the unsaponifiable matter containing the cannabinol dissolved in chloroform, the solution evaporated, the residue dried and weighed, and its iodine value determined. The following results were thus obtained with three samples of fresh charas and two samples several years old.1. 2. :3. 4. 3. (190s.) (1908.) (190s.) (1903.) (1893.) Unsaponifiable matter, per cent. 87% 84.4 83.5 66.4 54.8 Iodine value . . . ... ... 183 184 179 141 131358 Sucrose. THE ANALYST. Ash. Destrin. The iodine value of the dried extracts without previous purification also distin- guishes between old and fresh samples. From 0.1 to 0.2 gram of the extract is treated with 25 C.C. of freshly prepared Hiibl's solution, which is allowed to act for six hours.The following results were thus obtained : 1. 2. 3. Kashgar Charas, 1908 ... 234 231 233 Baluchistan Charas, 1905 , . . 202 205 183 (mouldy) ,, 1903 ... 196 202 Bhan; ... 1908 ... 173 (1905) 145 (1893) 112 Ganja ... 1908 ... 163 (1893) 130 C. A. 31. 'er Cent. 3.73 3.43 3.70 Composition of American Honeys. C. A. Browne and W. J. Young- (US. Dept. of Agric., Bull. 110, 1908, 1-94.)-The investigation recorded in this Bulletin was undertaken for the purpose of determining the general composition of American honey, and to establish a general range for the variation in the composition, which would be of use to the analyst in the examination of commercial samples. Analyses were made of about 100 samples, including 85 specimens from various States exhibited at the St.Louis Exhibition (1903), 11 samples from Hawaii, and 4 New England honeys. The average results obtained are given in the following table : Per Cent, 0.08 0.12 0.09 Ltevorotat ory honeys (92 samples) ... Dextrorotatory honeys (7:samples) All honeys (99 samples) ... ... 1.90 3.01 1.98 Lsvorotatory honeys Maximum ... Minimum ... Dextrorotat,ory honeys : Maximum ... Minimum ... 0.18 1.51 0.81 9.70 0.23 2.09 Water. 10.01 0.00 5.28 0.61 Per Cent. 17.70 16.09 17 -59 26.88 12.42 17-80 13-56 0.90 0.03 1.29 0.29 Invert Sugar. 'er Cent. 74.98 66.96 74-41 Per Cent, Per Cent. Per Cent. I EXTREME RESULTS. 83.36 62.23 71.69 64-84 7.58 0.04 12.95 6.02 I Qot-Esti- formic). 7.45 0-04 4-95 1-57 0-25 0.04 0.19 0.05 teciucing Sugars (as dex- trosih)..'er Ci,nt. 71.05 64.15 70.59 79.86 59.61 68.68 62-12 For detecting the presence of added invert sugar in honey, Ley's test (ANALYST, 1907, 32, 291) was found to be of some value, although it is not infallible. The greenish after-tone failed with many of the American honeys, but the brownish coloration of the reaction is useful as a confirmatory test. A positive reaction with aniline acetate may be regarded as an almost certain indication of adulteration withTHE ANALYST. 359 invert sugar. In this test a concentrated solution of the honey is placed in a test- tube and a solution of aniline acetate (prepared by mixing 5 C.C. each of aniline and water and adding sufficient acetic acid to clear the emulsion) is poured over its surface. If, when the tube is gently agitated, a red ring forms beneath the aniline acetate, this colour becoming gradually imparted to the whole layer, artificial invert sugar is present.The Bulletin also contains a description of the methods used in the microscopical examination of honey ; a number of photoinicrographs of different pollen grains are given, as well as a comprehensive list of the literature relating to honey. w. P. s. Calculation of Milk Adulteration. J. F. Liverseege. (Jozwn. SOC. Chem. I d . , 1908, 27, 604-605.)-1n calculating the composition of milk adulterated by the addition of both skimmed (or separated) milk and water, the usual method is by simultaneous equations, thus : 1. 0.085m + 0.09s = Non-fatty solids found in sample. 2. 0.035m + 0-002s =Fat found in sample.3. m+s+w==100, where m=percentage of average milk (taken as 3% per cent. fat and 8.5 per cent. non-fatty solids) ; s =percentage of Separated milk (containing 9.0 per cent. total solids and 0.2 per cent. fat) ; and w = percentage of added water. As the solution of such equations is tedious, the author has plotted out two curves, by means of which the composition of any sample may be readily calculated in terms of milk of average quality, separated or skimmed milk, and added water if any be present. The original paper should be consulted for reference to the curves and tables worked out by the author. A. R. T. Determination of Orcinol in “Orchella. Weed.” H. E. Watt. (Jozcrn. SOC. Chenz. Ind., 1908,27, 612.)-The author has modified Stenhouse’s original method (Jozcrn.p n k t . Chem., 1848, 45, lSOj, in which the colouring matter is extracted from the lichen by a solution of caustic alkali, and the liquid titrated with sodium hypochlorite solution which has been standardised against pure orcinol. Two grams of the powdered lichen are mixed in a mortar with an equal weight of sodium carbonate, and water gradually stirred in until the whole measures 100 C.C. The mixture is allowed to stand, with frequent stirring, for twenty minutes, when 50 C.C. ( = 1 gram lichen) is filtered off, excess of a standard solution of sodium hypochlorite added (which produces a fugitive blood-red colour), followed immediately by excess of a decinormal arsenious acid solution, The liquid is then titrated back with the sodium hypochlorite solution, using drops of acetic acid starch-iodide eolution on a porcelain tile as an indicator.It was found by experiment that, besides the instantaneous reaction which takes place between ( ( erythrin ” and the hypochlorite, a slow secondary action occurs. For this reason the excess of hypochlorite is immediately destroyed, and only that taken up instantaneously is measured. One decigram of orcinol was found to require360 THE ANALYST. 7.7 C.C. of a sodium hypochlorite solution, 4.55 C.C. of which were equivalent to 50 C.C. of decinormal arsenious acid solution. A. R. T. Estimation of Phosphoric Acid in Foods. E. Wormer. (&its. Unterszcch. Nahr. Genzmm., 1908, 15, 732-734.)-1n the method described the food material is oxidised by treatment with nitric and sulphuric acids, the phosphoric acid is precipi- tated with molybdate solution, and the precipitate is then heated with excess of standardised sodium hydroxide solution.The ammonia is thus expelled, and its quantity (as estimated by titrating back the excess of sodium hydroxide) is a measure of the amount of phosphoric acid present. I n the case of flour, from 1 to 5 grams of the sample are heated with 10 C.C. of a mixture of equal parts of concentrated sulphuric and nitric acids, a little more nitric acid being added from time to time until, when sulphuric acid fumes are given off, the solution does not darken in colour. The mixture is then diluted with water and filtered if necessary; to the filtrate, which should measure about 80 c.c., are added 30 C.C.of 50 per cent. ammonia; the mixture is heated to a temperature of 80" C., and the phosphoric acid is precipitated by the addition of 25 C.C. (or more, if necessary) of 10 per cent. ammonium molybdate solution. After fifteen minutes the supernatant liquid is poured through a small filter, the precipitate is washed by decantation with cold water until free from acid, using litmus-paper as indicator, and the filter is then placed in the vessel containing the precipitate. Water and a known volume of 2 sodium hydroxide solution are added, tbe mixture is boiled until all the ammonia has been expelled, and the excess of alkali is then titrated back, phenolphthalein being used as indicator. finder the above conditions the yellow phospho-molybdate precipitate has a constant composi- tion corresponding with the formula, 2(NH,),P0,.24M00,.4HN03.Each C.C. of c; alkali used corresponds with 0.00127 gram of phosphoric acid (P,O,), or 0*000556 gram of phosphorus. The process is applicable to the estimation of phosphoric acid in wine, beer, etc. w. P. s. Analysis of Turpentine Oil. F. W. Richardson and J. L. Bowen. (Journ. SOC. Clzern. Ind., 1908, 27, 613-616.)--Light hydrocarbons are almost the only adulterants of turpentine oil. Pine oils and pinolins, bye-products of the manu- facture of rosin oil, are powerful solvents of oils and resins, and as they are chiefly composed of volatile terpenes their use for ordinary purposes can scarcely be condemned, though they are objectionable for pharmaceutical use owing to their strong action on the skin.If the oil be shaken with half its volume of 6 per cent. iodine in carbon tetrachloride and heated for one minute at 100" C,, a dark green colour indicates the presence of pinolins (Valenta). Polarimetric methods are of little use as applied to turpentine oil, since varying amounts of lavo- and deztro-pinenes are present. Refractoinetric examination is useful, pure turpentine oil having a refractive index of 1.470 to 1473 at 15.5" C., while that of hydrocarbon adulterants varies from 1.410 to 1-460. The specijc refractivity (" 1, where n = refractive index and d = specific gravity is of little )THE ANALYST. 361 value in differentiating the pure oil from its adulterants. The speciific gravity of turpentine oil is a valuable guide as to its purity (average 0.864). Fractional distillation is one of the most useful methods o€ examination of turpentine oil, especially when combined with a refractometric determination of the fractions.The authors recommend, for general work, simple distillation of 100 C.C. of the oil in a, distilling-flask of 500 C.C. capacity, the thermometer-bulb being fixed just below the opening into the side-tube. Any residue over 200" C. should be transferred to a very small distilling-flask for final distillation. A sample of pure turpentine distilled as described gave 96.8 per cent. of distillate between 157" and 171' C. The method o j V&es (Journ. SOC. Chent. Inn., 1903, 22, 1106), which consists in determining the differences in refractivity of the fractions distilled in fifths (20 per cent.each fraction), was not found to yield any useful results. The acetic acid test of the British Pharmacopoeia was found to be unreliable, though if the test be applied in a manner similar to the well-known Valenta turbidity test, useful qualitative results can be obtained. In conjunction with the physical data described, the iodine absorption (Hiibl) is of great value. The bromine thermal value is not of much service in the examination of oil of turpentine, since the differences in rise of temperature with various mixtures of turpentine oil and petroleum products are too small. As petroleums give variable thermal values, the authors sought for some reagent which would be without action on the saturated hydrocarbons and upon such olefins as are usually present, while producing considerable heat constantly with terpenes.Antimony trichloride appeared to answer these requirements, and the test was carried out by immersing a thermometer with small spherical bulb in a platinum tube containing mercury, this apparatus being placed in a glass tube containing 1 gram of the finely powdered antimony trichloride. By means of a twisting action this substance was made to form a thin layer on the lower parts of the tube, and the whole apparatus was then introduced into a beaker packed with some non-conducting material (or a vacuum-tube may be employed). Two C.C. of the turpentine oil at the same temperature as the substances in this apparatus are added to the antimony chloride and the increase of temperature noted. This is the " antimony trichloride value." Petroleum gives no rise of temperature, while most genuine oils of turpentine give a value of about 47' G.One genuine oil examined, however, gave a rise of only 22.8" C,, in spite of the fact that it had an iodine absorption similar to another genuine sample that gave a rise of 47.2" C. The antimony trichloride value cannot, therefore, be implicitly relied on, and it is evident that a genuine turpentine may contain bodies reacting with iodine reagent and not to antimony trichloride. The authors do not reoommend any of the polymerisation methods, as they find that with nitric acid nitro-terpenes are formed which dissolve some of the petroleum, and, in addition, partial oxidation of the petroleum takes place ; while with sulphuric acid the polymerised bodies apparently dissolve some of the unpolymerised substances. A. R. T. Estimation of Water in Gluten. M. P. Neumann and P. Salecker. (Zeits. Unterszuh. Nuhr. Genussm., 1908, 16, 735-738.)-The cake of moist gluten obtained from flour in the usual way is pressed by means of the thumb over the bottom of a362 THE ANALYST. flat nickel basin, The basin and its contents are then placed in an air-oven at a temperature of 120" C. At this temperature the gluten becomes constant in weight in about three hours; some decomposition takes place, as is shown by the alteration in colour and in the decrease in solubility in 70 per cent. alcohol, but the alteration does not affect the weight of the gluten, as the results agree with those obtained by the method described by Bremer (ANALYST, 1908, 56), and with the percentages of moisture found when the gluten is dried under reduced pressure at a temperature of 95" C. w. P. s.