THE ANALYST. 45 ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOODS AND DRUGS ANALYSIS. The “Aldehyde” Value of Milk. R. Steinegger. (Zeit. Unters. Nahr. Genussm., 1905, x., 659-671.)-The addition of formaldehyde to milk causes an increase in the acidity of the latter, apart from that due to the acidity of the formaldehyde solution itself. The acidity increases with the quantity of formalde- hyde added until it reaches a maximum with the addition of 1.8 per cent., or about 5 per cent. of the ordinary 40 per cent. formalin solution. The difference between the original acidity of the milk (expressed in Soxhlet-Henkel degrees) and the acidity of the same after the addition of 5 or 6 C.C. of formalin is termed by the author the ‘‘ aldehyde ” value of the sample. This value for normal milks varies between 5.8 and 8.5.For the milk from single cows the value remains practically constant. In the case of colostrum the value rises as high as 17.3, but decreases as the milk becomes normal. The increase is observed with boiled milk, showing that the phenomenon is not due to the action of an oxidizing ferment. As the aldehyde value varies with the amount of casein and other proteids in the milk, the increase of acidity is undoubtedly caused by the chemical reaction between the formaldehyde and these proteids. The determination of the aldehyde value is ‘proposed as a means of detecting the presence of added water in milk, as this form of adulteration decreases the percentage of proteids in the milk, and, consequently, the aldehyde value.Abstraction of fat does not influence the value. w. P. 8. The Quantity of Water in .Canadian ‘‘ Domestic ” Butter. A. McGill. (Canadian Inland Revenue Department, Bull. 107, l-l2.)-By domestic butter is meant butter as found in the retail market, and intended for home consumption.46 THE ANALYST. Of 180 samples of this class of butter examined, only 1 contained an amount of water in excess of 16 per cent. Eight of the samples contained from 14 to 16 per cent., and in 144 samples the water was below 10 per cent. One of the samples con- tained borate preservative. w. P. s. Detection of Phytosterol in Lard in the Presence of Small Quantities of Paraffin, and the Determination of Small Amounts of Paraffin in Lard. E. Polenske. (Arbeit. Kaiserl. Gesmdheitsamte, 1905, vol.22, pp. 576-583.)-The sensitiveness of the phytosterol acetate test (ANALYST, 1902, xxvii., 94) is sometimes vitiated by the presence of traces of paraffin in the fat under examination. The paraffin becomes more and more concentrated as the acetate is recrystallized, until at last the melting-point obtained falls to that of the paraffin-that is, about 53" c. By taking advantage of the slight solubility of phytosterol in petroleum spirit, it is possible to remove the paraffin. The unsaponifiable residue obtained from 100 grams of fat in the usual manner is placed in a cylinder of about 8-C.C. capacity, and treated with 1 C.C. of petroleum spirit (boiling-point below 50° C.). After stirring up the residue with a glass rod, the cylinder is stoppered and placed aside for about twenty minutes.The whole is then poured on to a small pellet of cotton- wool placed in the stem of a funnel, and the residue and filter washed five times with petroIeum spirit, using 0.5 C.C. of the latter each time. The residue of phytos- terol is finally dissolved off the cotton-wool with a little ether, evaporated, and acetylated. Besides removing the paraffin, the above process concentrates the phytosterol in the residue, as cholesterol is considerably more soluble in petroleum spirit. For the determination of small quantities of parafin in fats the crude, un- saponifiable residue, from 100 grams of the fat is heated in a strong glass tube with 5 C.C. of concentrated sulphuric acid to a temperature of 105" C. for one hour (a mixture of 40 parts of glycerol and 60 parts of water boils at 104' to 105" C.).After cooling, the contents of the tube are shaken out three times with petroleum spirit, using 10 C.C. each time, and continuing the shaking for one minute. The extracts are united, washed a few times with a little water, the petroleum spirit is evaporated, and the residue weighed after being dried at 100" C. Should only traces of paraffin be present it is advisable to subject the residue to a second treatment with sulphuric acid. w. P. s. The Determination of Tannin in Wine. L. Kramszky. (&it. anal. Chenz., 1905, vol. 44, pp. 756-765.)-1t is shown that the method of Neubauer and Lowenthal gives not only the amount of tannin, but the sum of the tannin and colouring matters. Barth's colorimetric method has the same drawback, and is also not generally applicable.An amrnoniacal solution of zinc sulphate (or of nickel sulphate) yields quanfitafive results with solutions of pure tannin, which agree well with those obtained by the Neubauer-Lowenthal method, and at the same time does not pre- cipitate the colouring matter of wine. The reagent is prepared by dissolving about 25 grams of pure zinc sulphate in water, a2ding ammonia until the precipitate first formed redissolves, then an additional 300 C.C. of ammonia, and makingTHE ANALYST. 47 up the liquid to a litre. For the determination of the tannin 50 C.C. of a red wine or 100 C.C. of a white wine are rendered alkaline with ammonium hydroxide and heated, no notice being taken of any precipitate that may be formed.The liquid is then treated with 20 C.C. of the reagent and stirred until the zinc tannate separates, after which 300 C.C. of hot water are added, and the whole allowed to stand, The precipitate is repeatedly washed by decantation with hot, slightly ammoniacal water, and finally brought on to the filter (preferably a Gooch’s crucible containing asbestos), dried at 100” to 130” C., and weighed. It is then oxidized by means of a few drops of strong nitric acid, and ignited after evapora- tion of the acid. The diiference between the two weights gives the amount of tannin. The author gives a series of results thus obtained with solutions of pure tannin and with wines, including some to which had been added known quantities of tannin. Comparative determinations were made by the method of Neubauer and Lowenthal, the difference between the results in the case of red wines ranging from 0.02 to 0.08 gram.According to Neubauer, a deduction of 0.03 to 0.04 gram must be made for the colouring matters in dark-red wines when the permanganate method is used. As regards the question of gallic acid, the author considers tbat it is probably only present in traces in sound red wines, but that unsound wines may possibly contain considerably more. Experiments to determine the influence of gallic acid on the determination of the tannin by the zinc method have given the following results : (1) Gallic acid is not precipitated by ammoniacal zinc sulphate solution. (2) Wines to which a known quantity of gallic acid has been added give, on analysis by the method of Neubauer and Lowenthal, 0, higher value than corresponds to the sum of the tannin, colouring matter, and added gallic acid.(3) I n certain cases the presence of gallic acid prevents the precipitation of part of the tannin by ammoniaoal zinc sulphate solution, C. A. M. A Honey Substitute. G. Reiss. ( Arbeit. Kaiserl. Geszmdheitsamte, 1905, vol. 22, pp. 666-668.)-Under the name “Fruktin ” an article has recently been placed on sale as a honey substitute. I t consists of cane-sugar and a small quantity of tartaric acid, a little caramel being also present. According to the directions given with each packet of the article, 500 grams of the latter are heated to boiling with 150 grams of water, and the mixture then kept in a warm place for half an hour in o’rder to obtain a honey-like product. The author has himself prepared artificial honey by treating “Fruktin” in this way, and from it obtained the following analytical data : Specific gravity of a solution of 1 part in 2 parts of water, 1.11 ; invert sugar, 33.39 per cent.; cane-sugar, 41.63 per cent. ; ash, 0.024 per cent. ; polarization of a 10 per cent. solution observed in a 200-mm. tube, +4”35’; after inversion, -2’44’. The proportion of cane-sugar is, as will be seen from the above figures, about four times greater than is found in natural honey, whilst the percentage of ash is much lower. At the same time, a mixture of 1 part of ‘‘ Fruktin ” honey with 3 parts of pure honey would contain about 10 per cent. of cane-sugar, and such mixtures would only differ from natural honey in their low percentage of ash and in the presence of tartaric acid in them.w. P. s.48 THE ANALYST. Valuation of Saffron. A. Jonscher. (Zeit. byentl. Chem., 1905, xi., 444-447.)-The stigma of saffron possesses a considerably higher aromatic or spice value than the pistil, as it contains more ethereal oil and crocin. The stigma is also superior to the pistil as regards colouring power. An approximate method for determining the value of a sample of saffron consists in boiling 0.1 gram of the air- dried sample with 10 C.C. of 50 per cent. alcohol, at once cooling the mixture, and then allowing it to stand for one hour. The solution is then filtered, and 5 C.C. of the filtrate diluted to 100 C.C. with water. The coloration is then compared with that produced by a good specimen of saffron under similar treatment.W. P. S. Determination of Vanillin. J. Hanus. (Zeit. Unters. Nahr. Genussm., 1905, vol. 10, pp. 585-591.)-Vanillin is quantitatively precipitated from its aqueous solution by m-nitrobenzhydrazide. The presence of the ordinary adulterants of vanillin, such as acetanilide, benzoic acid, salicylic acid, and sugars, is without influence on the precipitation. For the determination of vanillin in vanilla the following method, based on the above-mentioned fact, is described : About 3 grams of the vanilla, in small pieces, are extracted for three hours with ether. The ethereal solution is evaporated at a temperature of 60" C., the residue is dissolved in a little ether and filtered into a small flask, the filter being washed with ether.The ether is again evaporated, the residue heated with 50 C.C. of water for thirty minutes on the water-bath, and the emulsion so obtained precipitated with a solution of 0.2 gram of m nitrobenzhydrazide in 10 C.C. of hot water. The flask and its contents are placed on the water-bath for thirty minutes, and then set aside for twenty-four hours. The mixture is now shaken out three times vith petroleum spirit to remove fat, the extracts, after separation, being filtered in a Gooch's crucible, and the precipitate itself then rinsed on to the filter, washed first with water, then with petroleum spirit, dried at a temperature of 100" to 105" C., and weighed. The weight of the precipitate mdtiplied by the factor 0.4829 gives the amount of vanillin present.The method is also applicable to the determination of vanillin in alcoholic extracts, but the alcohol must first be removed by evaporation at a low temperature. w. P. s. The Determination of Iodine in Iodated Thymols. H. Cormimboeuf. (Ann. de Chim. anaZ., 1905, vol. 10, pp. 453-454.)-Iodated thymols (aristols) contain from 15 to 45 per cent. of iodine, according to the mode of manufacture. Most of the commercial products are chlor-iodo- rather than di-iodo thymols, and there is a risk of the whole of the halogen present being determined as iodine if the ordinary methods of Carius, etc., be used. The following method is recommended as giving only the iodine and requiring but little material : 0.5 gram of the sample is thoroughly mixed in a mortar with 3 grams of dry sodium carbonate, and the mixture heated gradually in a platinum or nickel crucible until nearly complete combustion of the organic matter, after which it is fused.The mass, when cold, is treated with hot water, the solution filtered from any particles of unburnt carbon, the filtrate mixed with half its volume of ammonium hydroxide, and the iodine precipitatedTHE ANALYST. 49 with silver nitrate in the usual way. precipitate of silver chloride being obtained in the presence of chlorine. The filtrate is acidified with nitric acid, a C. A. M. Oil of False Savin. ' J. C. Umney and C. T. Bennett. (Pharm. Jozcrn., 1905, vol. 75, pp. 827-829.)-The oil of savin as distilled in this country, in Germany, and in the South of France, varies considerably.This is probably due to the fact that in the latter country the oil is distilled from false savin, Juniperus Phmzicia (the common Mediterranean Apecies), whilst in the former the distillers use Juniperus sabina. The following results were obtained with three specimens of the oil : Specific gravity Optical rotation Esters ... ... Total sabinol ... Solubility in 90 alcohol .,. Portion distilling : Below 155" C. ,, 165" C. ,, 175" c. 7 , 180" c. ,, 200" c. ,, 220" c. ,) 230" C. ... ... ... ... ... ... ... ... per cent. ... ... ... .., ... ... ... ... ... ... ... ... ... ... .*. ... English. From J. Sabinn. 0.909 + 68" 47.6 per cent. 52-1 ,, 1 in 1 nil. nil. 14 per cent. 23 9 9 48 7, 62 9 , 80 9 , German. Fro111 J . Subina.0.920 + 42" 36.5 per cent. 48.2 ,) 1 in 1 nil. nil, nil. 4 per cent. 29 9 7 49 9 7 60 $ 9 French. From J. Phceniciu. 0.892 + 4O30' 9.3 per cent. 17.1 ,) 1 in 5 The oil from J. Phcenicia contains over 75 per cent. of pinene, the hydrochloride of which melts at 125O C., and the nitrosochloride, after one recrystallization, at 1 0 7 O C. A sesquiterpene, cadenine, is also present in the oil. w. P. s. Examination and Assay of Aloes. L. v. Itallie. (Pharnz. Weekblad, xlii-, 553; through Pharm. Journ., 1905, vol. 75, p. 554.)-The author considers that Tschirch's method (ANALYST, 1905, xxx., 249) is open to objection, and modifies it as follows : Five grams of the powdered aloes are warmed with 5 C.C. of methyl alcohol until a homogeneous liquid is obtained; 30 C.C.of chloroform are then added, the mixture is shaken for five minutes, and allowed to stand until clear. The solution is decanted, evaporated to dryness, and the residue again dissolved in methyl alcohol and precipitated. By the use of this method Cape aloes mere found to contain from 18 to 43 per cent. of resin, and Curacao aloes from 11 to 21 per cent. Objection is also raised to Tschirsh's pro- posed colorimetric determination of aloin by Schouteten's reaction (green fluorescence with borax solution) on the ground that it is far too subjective, the author finding 100 per cent. of aloin in a sample of aloes, Determination of aloin by precipitation, as tri-bromo-aloin, succeeded with pure aloin, but indicated 71 per cent. of aloin in This treatment is repeated a third time.50 THE ANALYST.Cape aloes and 72 per cent. in Curacao, which results the author considers to be much too high. w. P. s. Analysis of Powdered Extract of Nux Vomica. W. H. Lenton. (Phar??~. Journ., 1905, vol. 75, p. 864.)-The following process is particularly applicable to the analysis of the powdered extract of nux vomica, and is similar to Bird’s method (Pharm. Jozcrn., xi., 4, 574), dilute alcohol being used as a solvent for the powder, and a mixture of ether and chloroform for the actual extraction. Ten C.C. each of ether and chloroform are placed in a dry separating funnel, 2 grams of the powder are introduced, then 5 C.C. of 90 per cent. alcohol and 3 C.C. of concentrated solution of ammonia, are added, and, finally, 5 C.C. of water. The whole is well shaken for about one minute, allowed to settle, and the ether-chloroform layer drawn off into another separating funnel, where it is shaken with 5 C.C. of 10 per cent. ammonium carbonate solution. After shaking the ether-chloroform once more with ammonium carbonate solution, the former is drawn off and placed aside. The alkaline mother- liquor is twice again shaken out with 20 C.C. of the ether-chloroform mixture, and the extract washed with the ammonium carbonate used for the first portion drawn off. The mixed ether-chloroform solutions are then treated as usual for the determination of strychnine. w. P. s.