THE ANALYST. 21 ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOODS AND DRUGS ANALYSIS. Observations on the Analysis of Food Materials. H. Leffmann. (Chem. Zeit., 1905, xxix., 1086.)-I. Detection of Abrnstol (AsaproZ).-Abrastol is a P-naph- tholsulphonate ; it has antiseptic properties and can be used as a food preservative. The author proposes the following process for its detection in milk : Ten C.C. of milk are treated with 0.5 C.C. of a solution of acid mercuric nitrate (prepared by dis- solving mercury in twice its weight of nitric acid and diluting with five times the quantity of water). If abrastol is present a yellow coloration is quickly shown. A control experiment with pure milk should be made at the same time. In wines, etc., abrastol may be detected by acidifying with a few drops of dilute sulphuric acid and shaking with an equal volume of ether, chloroform, etc.The solution is drawn off, and a, few drops of the mercuric nitrate solution added and the liquids shaken22 THE ANALYST. Abrastol is indicated by the mercury solution turning yellow, and finally bright red. Benzoates and salicylates give no reaction in this test. 11. Estimation of Methyl Alcohol in Presence of Foyma1dehyde.-The formalde- hyde is first removed by addition of potassium cyanide solution ; sufficient cyanide has been added when a sample gives no blue colour on adding phenylhydrazine hydrochloride, some very dilute, freshly-prepared nitroprusside solution, and then sodium hydroxide (phenylhydrazine nitroprusside test). The liquid is then distilled, and the methyl alcohol estimated in the distillate by conversion to formaldehyde in the usual way.The method gives much better results than the ammonium hydroxide or hydrogen peroxide processes. E. K. H. A Volumetric Method of Estimating the Cinchona Alkaloids by means of their Double Thiocyanates. P. W. Robertson. (PTOC. Chem. SOC., 1905, xxi., 242.)-Many alkaloids give precipitates with ammonium thiocynate in the presence of a zinc or mercury salt. Other metallic salts act in a similar manner, but zinc forms the most insoluble precipitate. The alkaloids most sensitive to this reaction are quinine and the cinchona alkaloids, 1 part of quinine in 50,000 parts of water giving a distinct turbidity when treated with an excess of zinc sulphate and ammonium thiocyanate. These precipitates prove to be double salts of groat complexity, cinchonine ammonium zinc thiocyanate having the formula : 4C,,H,,ON,.3Zn( CN S),.2NH4CN S.,LHC& S., which closely resembles that of herapathite or iodoquinine sulphate : 4C,oH,,0,N2.3H,S04.2HI.41,.3Aq. The determination of the amount of thiocyanate removed from solution by the alkaloids is said t o form an accurate and rapid volumetric method of determining quinine in drugs and for the assay of crude cinchona bark. w. P. s. Assay of Calabar Beans. Beckurts. (~?poth. Zeit., XX., 670 ; through Phtarm. Journ., 1905, vol. 75, p. 584.)-Physostigmine can be completely extracted from an aqueous solution by ether, and the following method is proposed for the determination of the ether-soluble alkaloid present in the seed; the extracted alkaloid being subsequently titrated, using iodeosine as indicator.Twenty grams of the seed are shaken for three hours with 120 grams of ether and 10 C.C. of 10 per cent. potassium hydrogen carbonate solution. Ninety grams of the clear ether are then poured off, distilled to one-half, transferred to a separating funnel, and shaken out several times with small quantities of & hydrochloric acid, 10 C.C. of petroleum spirit being added to prevent the formation of emulsion. To the united acid extracts 45 grams of ether and 10 C.C. of 10 per cent. potassium hydrogen carbonate are added, and the mixture well shaken. After separation has taken place, 30 grams (equal to 10 grams of seed) of the ether are decanted off; 10 C.C.of -& hydrochloric acid, 20 C.C. of water, and 5 drops of an alcoholic solution of iodo-eosine are added, and the excess of acid titrsted back with i& sodium hydroxide. The method is also applicable to the extract. The author found 0.0825 per cent. of alkaloid in the seeds, and from 1-25 to 1.3 per cent. in the extract. w. P. s.THE ANALYST. 23 Assay of Coca Leaves. Greshoff. (Apoth. Zeit., 1905, XX., 291; through Plzarrn. Joum., 1905, vol. 75, p. 724.)-The following method is employed at the Haarlem Colonial Museum for determining the value of Java coca leaves: 30.5 grams of the dry, finely-powdered leaves are heated on a water-bath a t about SO" C. under a reflux condenser for two hours with 300 C.C. of alcohol, the weight of the flask and contents having been noted.After cooling, the quantity of alcohol lost is made up, and 150 C.C. (equivalent to 15 grams of leaves) are filtered off and evaporated, with constant stirring, nearly to dryness. When cold, the insoluble residue is collected on a filter and washed with tepid water until about 60 C.C. of filtrate have been collected. This filtrate is shaken out twice with 30 C.C. of ether, which are rejected ; the aqueous solution is then rendered alkaline with ammonia, and extracted with three successive quantities of 30 C.C. of ether. The united ethereal extracts are evaporated, and the residue dried in a water-oven, a current of dry air being meanwhile drawn through the flask to remove volatile base with a strong tobacco-like odour. The amorphous straw-yellow residue is then dissolved in a little 1 per cent.sulphuric acid; the acid solution is again made alkaline with ammonia and the extraction with ether repeated. The residue obtained on evaporating the ether is dried for three hours and weighed. The following quantities of alkaloid were obtained by the above process in various samples of leaves : Young leaves, 2.02 per cent. ; old leaves, 0.78 per cent. ; selected apical leaves, 2.1 per cent. ; leaves from the base of the plant, 1.2 per cent. The author considers that commercial samples of the leaves should not contain less than 0.6 per cent. of the alkaloid, calculated on the dry substance, w. P. s. The Determination of Iron in Ferrum Redactum. H. Cormimboeuf and L. Grosman. (Ann. de Chim. anal., 1905, vol.10, p. 420-422.)-The following modi- fication of the iodine method is recommended as simpler and more reliable than that given in the German Pharmacopoeia: 1 gram of the reduced iron is treated with 25 C.C. of a double normal solution of iodine (254 grams of iodine, and 360 grams of potassium iodide per litre), and the mixture allowed to stand for at least six hours with occasional agitation, after'which a large excess (250 to 300 c.c.) of water is added, and the residual iodine titrated with standard thiosulphate solution. Each C.C. of the iodine solution corresponds to 0.056 gram of iron (cf. ANALYST, xxx., 338). C. A. M. Assay of Sublimate Gauze. Utz. (Plzwm. Post., xxxviii., 491 ; through Phamt. JozLm., 1905, vol. 75, p. 691.)--It is well known that in sublimate gauze a part of the mercury enters into combination with the cellulose and cannot be extracted by solvents. The following method, however, originally devised by Rupp, gives excellent results in the determination of mercury in this article : Five grams of the material are placed in a stoppered flask, together with sufficient water to saturate it, and 10 C.C. of a mixture of equal parts of formaldehyde solution and 15 per cent, sodium hydroxide solution are added. After heating for fifteen minutes on a water-bath, the contents of the flask are cooled, 5 C.C. of glacial acetic acid and 5 C.C. of & iodine solution are added, and the whole allowed to stand for some time with occasional24 THE ANALYST. agitation. The excess of iodine is then titrated back with thiosulphate solution, care being taken to close the flask and shake vigorously towards the end of the reaction ; 1 C.C. of Tv iodine indicates 0.01355 gram of mercuric chloride. w. P. s.