72 THE ANALYST. ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOODS AND DRUGS ANALYSIS. The Detection, Determination, and Rate of Disappearance of Form- aldehyde in Milk. R. H. Williams and H. C. Sherman. (Journ. Amer. Chem. SOC., 1905, xxvii., 1497.)-The authors find that by means of the potassium cyanide method (ANALYST, 1904, xxix., 5) formaldehyde in milk can be approximately determined at any concentration greater than 1 part in 160,000. They find that aqueous solutions of formaldehyde containing from 1 : 5000 to 1 : 40,000 lose steadily in strength on standing, the loss being not merely due to polymerization, but to actual destruction of the formaldehyde. Within the same limits, formaldehyde added to milk disappears about ten or twenty times as rapidly as from aqueous solutions of the same strength.Stronger solutions are much more stable; thus, an aqueous solution of 1 : 1,000 showed no appreciable loss after five months. From milk containing 1 part in 1,000 the rate of disappearance was the same as for an aqueous solution of 1 : 5,000. Souring of the milk does not affect the rate of dis- appearance. The hydrochloric acid and ferric chloride test is sensitive to a dilution of 1 : 250,000. The test may be made more satisfactory by adding 50 to 75 CA. of cold water to the milk after keeping it just below the boiling-point for one minute as usual; the colour obtained in this case, however, is very fugitive. The gallic acid test is much more delicate than the ferric chloride test. Sourness of the milk does not affect the reaction. A.G. L. Composition of the Fat of Pigs fed on Foods Rich in Oil. K. Farn- steiner, K. Lendrich, and P. Buttenberg. (Zeit. Untersuch. Nab. Gemssm. , 1906, xi., l-S.)-The particular object of the present experiments was to ascertain whether the oily constituents of the fodder were deposited in the body fat of pigs fed on such fodder. The results show that the fat of pigs fed on cottonseed-meal gives a decided reaction with Halphen's reagent, but that the fat in no case contains phytos- terol, and is consequently thus distinguished from lard adulterated with cottonseed- oil. (Cf. ANALYST, 1904, p. 315.) w. P. s. On the Occurrence of Arsenio in Wines. H. D. Gibbs and C. C. James. (Journ. Amer. Chem. SOC., 1905, xxvii., 1484.)-The authors have exa.mined a total of 352 wines, some bottled and some in cask, and have found arsenic in 52 cases, in quantities ranging up to 1 part of arsenic in 2,500,000 of wine.Generally, however, the quantities found were much smaller than this. As probable sources of the arsenic they indicate the arsenical sprays used upon the vines, the sulphur burned for sulphuring the wine and receptCtcles, and, perhaps, also the lead-shot used in cleaning the bottles. The method used in testing for arsenic consisted in making the wine alkaline with milk of lime, evaporating to dryness, and igniting the residue a t a, low red heat. The ash was treated with sulphuric acid (1 to 3), the solution filteredTHE ANALYST. 73 and introduced into the generator of a Marsh apparatus. As generator a 250 C.C.Florence flask was used, Kahlbaum's zinc and pure sulphuric acid being used to generate the hydrogen. The action was started by adding a few drops of platinic chloride.':: Immediately after adding the sulphuric acid the air in the apparatus was swept out by a rapid current of carbon dioxide, and the solution to be tested was then added. The evolved gas was passed through a spiral of lead acetate paper and then through calcium chloride, after which it entered a Jena glass tube of 0-7 om. internal and 1.0 cu. external diameter drawn down to a constriction from 10 to 15 em. long and 2 mrn. external diameter. The tube was heated for a space of 10 to 15 cm., pro- tected by wire gauze, before the constriction, the heating being continued for one hour. The electrolytic generator of Thorpe was also tried, but discarded because in some eases it failed to reveal the presence of any arsenic in wines known to contain arsenic.A. G. L. Process for Determining the Quality (Fineness) of Flour. N. Wender. (Zeit. U?ztersz~ch. Nahr. Gem~ssm., 1905, x., 747-756.)-Wheat and other grains contain an enzyme capable of liberating oxygen from hydrogen peroxide, and upon the fact that this enzyme resides principally in the embryo and outer portions of the grain, the author bases a process for distinguishing between best white flour and other flours containing more of less bran or inferior flour. I t is recognised that;, taken by itself, the colaur of flour is no criterion in judging whether a sample is best quality, " seconds," '' thirds," etc.The process is carried out as follows : 25 grams of the flour are rubbed down with 100 C.C. of water until free from lumps and then rinsed into a flask with a further 100 C.C. of water. The flask is closed by an indiarubber stopper carrying a tapped funnel and a delivery-tube connected to a measuring-tube filled with water. Ten C.C. of hydrogen peroxide are then introduced through the funnel, and the action allowed to proceed for about thirty minutes at a temperature of 20" C. Tho volume of the liberated gas is then read off. The following examples are given, in which the results are calculated for 100 grams of meal : Wheat starch, 8 C.C. ; wheat flour, 169 C.C. ; wheat bran, 342 C.C. ; rye flour, 153 C.C. ; rye bran, 330 C.C. ; maize flour, 389 C.C. In a series of flours of decreasing quality, the finest flour gave 128 c.c., and the coarsest 486 C.C.The author admits that many more analyses must be made before any degrees of quality can be fixed by this method, but considers that the latter will be of use when the flours of different countries, mills, etc., have been examined and classified. w. P. s. A New Adulterant of Lemon-Juice. H. Matthes and F. Muller. (Zeit. Untersuch. Nahr; Genzhssm., 1906, xi., 20, 21.)-A sample of so-called lemon-juice gave the following results : Alcohol, by weight, 5.43 per cent.; total solids, 13.47 per cent.; total acidity, as citric acid, 8.19 per cent.; mineral matter, 0.205 per cent.; alkalinity .of ash, 0.49 C.C. acid ; phosphoric anhydride, 0.05 per cent. ; polarization after distillation, + 4.5" in 100 mm.tube. The sample contained starch-syrup, evidently * It has been shown that the use, in this way, of platinic chloride is inimical to the cosrect estima- tion of arsenic. cf. Chapman, this vol., p. Q.-EDITOR.74 THE ANALYST. Alcohol ... ... ... ... I 14.55 Total solids ... ... ... ... 10.11 Mineral matter ... .., 1.. 0.065 Total acidity, as tartaric ... 8.029 Sugar ... ... ... ... ... none added to imitate the pectinous substances of pure lemon-juice and the amount of phosphoric anhydride found was probably due to the addition of a salt of this aoid. w. I?. s. 19.40 19.91 27.84 11.60 18.54 9.74 11.56 15.57 9.38 0.174 0.082 0*119 none none none Composition of Lemonade Essences. Utz. (Zeit. ofentl. Chem., 1906, xii., l2-13.)-The liquids to which the following analyses refer are sold for the purpose of making lemonade and similar drinks by diluting the liquid with water, soda-water, etc.They consist essentially of an aqueous solution of tartaric or citric acid, to which an alcoholic solution of certain flavouring substances (esters) are added. The figures exprese grams per 100 C.C. : Raspberry 1 Raspberry 1 Lemon 1 Lemon Essence A. Essence B. Essence A. Essence B. All four samples were strongly coloured with coal-tar dyes, and two of them- Raspberry Essence B and Lemon Essence B-contained not inconsiderable quantities of saponin. w. P. s. On the Colourihg Matter of Saffron. F. Deeker. (Chem. h i t . , 1906, xxx., 18.)-Crocetin, the colouring matter of saffron, has hitherto only been isolated in an amorphous form, since the readiness with which it resinified prevented its being obtained in 8 crystalline condition.The author, however, has succeeded in obtaining crocetin crystals of constant composition by digesting the resin-free solution of the colouring matter in very dilute sodium hydroxide solution at 60" to 70" C., with an excess of ammonium carbonate solution. The ammonium compound of crocetin was deposited from the liquid on cooling, in the form of yellow needles, which dissolved readily in alkalies, but were only sparingly in water and alcohol. The addition of ammonium carbonate to any of these solutions reprecipitated quantitatively the crocetin compound. An elementary analysis gave the following results : Carbon, 64.60 ; hydrogen, 8-70 ; nitrogen, 8.06 ; and oxygen, 18.64 per cent.C. A. M. Separation of the Conium Alkaloids. J. von Brawn. (Berichte, 1905, xxxviii., 3108-3113 ; through Pharm. Jour., 1905, vol. 75,909.)-The separation of the alkaloids of the hemlock-namely, coniine, methyl-coniine, y-coniceine, conhydrine, and pseudo-conhydrine, may be effected as follows : The greater portion of the coniine, which forms the chief constituent, is separated by distillation and the residue then fractionated up to 190" C., whereby the conhydrine (m.p. 118" C.) remains as an undistilled residue. The lower fractions are next benzoylated in alkaline solution, and the resulting oil, after dissolving in ether, is extracted with acid to remove therTHE ANALYST. 75 methyl-coniine. The ethereal solution is then concentrated and treated with petroleum spirit (ligroin) to precipitate the benzoyl-amino-butyl-propyl-ketone, which is formed by the action of benzoyl chloride on y-coniceine.The mother-liquor is redistilled and yields benzoyl coniine (b.p. 203'to 204" C. at 16 mm. pressure), and some more of the benzoyl-amino-ketone. Starting with 104 grams of material, the author was able to obtain 1 gram of conhydrine, 7 grams of methyl-coniine, 52 grams of amiho-ketone corresponding with 26 grams of y-coniceine, and 124 grams of benzoyl- coniine, corresponding with 68 grams of coniine, total 102 grams. The coniine recovered from the benzoyl derivative proved to be a mixture consisting of t-coniine and a little d-coniine. The substance described above as methyl-coniine has [.ID 35.66 at 24' C., and appears to be a simple substance.I t yields a picrate melting at 114" C., a platinichloride melting at 195" C., and an aurichloride between 80" and 90" c. w. P. s. The Oleoresin of Pinus Longifolia. F. Rabak. (Pharm. Rev., 1905, xxiii., 229 ; through Phurm. Jour., 1906, vol. 76, 11.)-The oleo-resin of Pinus Zongijolia, a native of the lower Himalayas, is a white, opaque, tough substance showing a partially crystalline structure. A sample examined had a specific gravity of 0.866 and [a],, + 2" 48. I t yielded 18.5 per cent. of essential oil on distillation, the oil having a characteristic pinene odour, and a faint aroma of limonene. On fractionation, dextropinene and laevolimonene were isolated. The residual resin was whitish and brittle ; it had an acid value of 142 and ester value of 13.When dissolved in acetic acid the solution deposited crystals of a resin acid Its odour is lemon-like. melting at 138' to 140" C. w. P. s. A Distinctive Reaction for Euphorbium Resin. A. Tschirch and Paul. (Pharm. Zeit., 1905, l., 561 ; through Pharm. Joum., 1906, vol. 76, 35.)-About 0.01 gram of the resin is treated with 10 C.C. of petroleum spirit, and the mixture filtered. The filtrate is allowed to flow on to the surface of 20 C.C. of sulphuric acid, to which 1 drop of nitric acid has been added. A red-coloured zone is obtained at the junction of the two liquids; on shaking, the acid layer becomes red and, in the course of a day or two, turns brown. w. P. s. Estimation of Formaldehyde in Formaldehyde Pastilles (Trioxy- methylene).Ernst Rust. (Zeit, angew. Chem., xix., 138.)-The author claims the following method of estimating formaldehyde to be superior to the older ones : Two grams powdered trioxymethylene are added to a 250 C.C. Erlenmeyer flask, in the neck of which is fitted a funnel. The solid is washed in with 70 C.C. normal sodium hydroxide solution. To this mixture 9 to 10 grams of 30 per cent. hydrogen peroxide is added slowly at first to prevent undue heating. After standing for two hours the mixture is warmed gently to boiling, to decompose the excess of hydrogen peroxide. The funnel, after being carefully washed with water, is removed, a small excess of normal sulphuric acid solution added, and the mixture titrated back with normal sodium hydroxide solution, using phenolphthalein as indicator. One C.C. of normal sodium hydroxide corresponds to 0.03 gram CH,O. The original substance76 THE ANALYST+ must be tested for alkalies or acids before this method is applied, and these must be taken into account in the final calculation. H. D. L. The Detection of Iodine Compounds in the Dry Way. B. Merk. (Pharm. Zeit., 1905, l., 1022; through Chem. Zeit. Rep., lT5, xxix., 405.)-The substance under examination is rubbed with a little potassium persulpbafe and some soluble starch, the mixture becoming more or less blue in the presence of iodine compounds not containing oxygen : 2KI + K,S20s = 2K2S04 + I,. C. A. M.