FOOD AND DRUGS ANALYSIS 273 ABSTRACTS OF PAPERS PUBLISHED IN OTHER JOURNALS. FOOD AND DRUGS ANALYSIS. Estimation of Small Quantities of Alkaloids. E. Carlinfanti. (Boll. Cliim. Fawn., 1915, 54, 321-323; through J. Chem. Xoc., 1915, 110, ii., 709-710.)- The estimation of small amounts of morphine or codeine, either in pharmaceutical preparations or for toxicological purposes, may be effected by the following colori- metric methods, a Wolff colorimeter being employed : In the case of morphine, the solution to be examined and a known volume, say 1 C.C.of a 0.5 per cent. morphine hydrochloride solution, are evaporated separately in basins on a water-bath, and allowed to cool in a desiccator. Each of the residues is dissolved in 5 C.C. of con- centrated sulphuric acid, and the liquid introduced into a tube holding about 50 C.C.and fitted with a ground stopper; each dish is washed twice with 3 C.C. of con- centrated sulphuric acid, and the washings added to the tube, which is then closed and immersed for fifteen minutes in a boiling water-bath. To the cooled tubes are added 10 C.C. of a Ilzixtcrp, of 100 C.C. cf cencectratei? sdphuric acid, with 2 d r ~ p s of nitric acid (sp.gr. 1.4); on shaking the tubes, the characteristic blood-red coloration appears. The solutions are then introduced into the cylinders of the colorirneter, in which they are made up to the same height by addition of the small quantities of concentrated sulphuric acid used for washing the tubes. In the case of codeine, to each of the two aqueous residues, reduced to about 1 C.C.by evaporation on a water-bath a t 70" to 75" C., 15 to 20 C.C. of monohydrated sulphuric acid are added cautiously, so that the mixture is not heated. The liquid is then introduced into a 50 C.C. flask, together with three quantities of 5 C.C. of the acid used to wash out the dish and 10 C.C. of monohydrated sulphuric acid containing 2 C.C. of 10 per cent.ferric chloride solution per 100 C.C. of the acid. After being shaken, the flasks are immersed for fifteen minutes in a water-bath at 80" C., the cooled solutions, which have acquired blue colorations, being introduced into the cylinders of the colorimeter as before. These methods admit of the estimation of 0.001 to 0.003 grm. of morphine or codeine hydrochloride to within about 0.0001 grm.Estimation of Small Quantities of Alkaloids. C. Carlinfanti and M. Scelba. (Boll. Chim. Farm., 1916, 55, 225-232; through J. Chem. SOC., 1916, 110, ii., 356-357.)-Under the conditions given for the estimation of morphine (see preceding abstract), acetoxy-morphine (heroine) may be estimated with considerable accuracy ; it may be distinguished from morphine, since it colours concentrated sulphuric acid containing nitric acid orange-yellow in the cold and blood-red when hot, and does not reduce iodic acid.Codeine may also be estimated similarly, but for very small quantities of this alkaloid the method previously given (Zoc. cit.) is recommended.274 ABSTRACTS OF CHEMICAL PAPERS Apomorphine may be estimated as follows : 1 to 5 C.C.of a 0.1 per cent. solution of the hydrochloride is evaporated to dryness in a basin on a water-bath. The residue is allowed to cool in a desiccator, and 10 C.C. of 95 per cent. alcohol and 0.1 grm. of sodium bicarbonate immediately added. The whole is then covered with a watch- glass and stirred repeatedly during four to five hours, after which the liquid is poured into a, tared flask and made up to volume with the washings of the residue with 95 per cent.alcohol. When the suspended sodium bicarbonate has settled, a portion of the clear liquid is decanted into a colorimeter, and the emerald-green coloration which develops compared with that obtained from a known quantity of the alkaloid treated similarly. This method gives good results with quantities of apomorphine of the order 0.001 to 0.002 grm.For smaller amounts, the authors recommend the fol- lowing modification of Grimbert and Lecldre’s method (ANALYST, 1915, 40, 121) To the solution of the alkaloid, made up with water to 3 c.c., are added 5 drops of saturated mercuric chloride solution and then 5 drops of 10 per cent. sodium acetate solution. The liquid is then boiled for half a minute, cooled, mixed with 1 C.C.of amyl alcohol, and introduced into a 50 or 100 C.C. flask, the vessel being washed out several times with small quantities of 95 per cent. ethyl alcohol, and the volume mado up with concentrated alcohol. After being shaken, the liquid is left until the mercurous salt settles, the clear solution being compared in the colorimeter with one prepared similarly from a known weight of apornorphine.I n the case of strychnine, a definite volume of the solution containing at least 0.004 to 0.005 grm. of the alkaloid is heated to boiling with 20 to 25 C.C. of 15 per cent. sulphuric acid solution, recently prepared bromine water being added, drop by drop, until the liquid assumes a pale yellow colour, and the boiling then con- tinued for a few minutes; the presence of strychnine is revealed by a more or less intense reddish-violet coloration.Further addition of a few drops of bromine water turns the hot acid liquid pala yellow, and subsequent boiling renders it reddish- violet again. The cold solution is made up to 50 to 100 c.c., and its depth of colour matched with that of a solution prepared similarly from a known weight of the alkaloid.With a pharmaceutical preparation containing a strychnine salt, a quantity containing at least 0.005 grm. of the base is mixed with an a!kali, and the mixture extracted with chloroform, the residue from the latter then being dissolved in 15 per cent. sulphuric acid solution and treated as above. Under the conditions employed in the case of shrychnine, brucine gives a salmon- red coloration with bromine water, but, when the alkaloid is present in small propor- tion, the second addition of bromine water renders the solution colourless.With strychnine, on the other hand, the violet-red coloration persists almost unaltered after a second and even a third treatment with bromine water. I n conlormity with this behaviour, the proportion of strychnine in Nux vomica preparations and other solutions containing strychnine and brucine in approximately equal proportions may be estimated by means of bromine water in the manner described above.Colorimetric Determination of Cinnamaldehyde in Cinnamon. T. von Fellenberg. (Mitt. LebensmittelzLizters. Hyg., 1915, 6 , 254-266 ; through J. Chem. Soc., 1916, 110, ii., 354-355.)-The value of cinnamon lies in its cinnamaldehyde contentFOOD AND DRUGS ANALYSIS 275 rather than in the quantity of total essential oil it contains.A method for the estimation of cinnamaldehyde depends on the coloration which develops when the aldehyde is treated with sulphuric acid and isobutyl aicohol. One grm. of the cinnamon is heated just to boiling for ten minutes with 40 C.C.of 95 per cent. alcohol in a flask attached to a condenser, and any distillate collected in a 100 C.C. flask. From 30 to 35 C.C. of the alcohol are then distilled, 100 C.C. of boiled water are added to the residue, and the distillation is continued until the total distillate measures 100 C.C. ; 5 C.C. of the distillate are then mixed with 2 C.C. of 5 per cent.isobutyl alcohol solution (in 95 per cent. alcohol) and 3 C.C. of 38 per cent. alcohol, 20 C.C. of concentrated sulphuric acid are added, and, after forty-five minutes, the coloration obtained is compared with that yielded by a known amount of cinnama€dehyde under similar conditions. The standard cinnarnaldehyde solution used for comparison contains 2 per cent. of the aldehyde in 38 per cent.alcohol solution ; it may be prepared from the aldehyde-sulphite compound, and standardised by a bromine-iodine titration. Nine samples of Ceylon cinnamon were found to contain from 1.31 to 1.84 per cent. of cinnamaldehyde ; seven samples of cassia-cinnamon contained from 1.23 to 2.77 per cent., and a sample of cinnamon flowers 3.73 per cent. Colour Reaction of Croton Oil.Comte. ( J . Plzarz. Chim., 1916, 14, 38-39.) -Croton oil contains an active resinous constituent, which, like the oil itself, is The oil under examination is treated with twice its volume of absolute alcohol, and a little of the clear alcoholic solution poured on to a concentrated solution of potassium or sodium hydroxide. A brilliant brownish-red or reddish-violet ring ,(according to the age or origin of the oil) indicates the presence of croton oil.This reaction is not given by olive, poppy, sesame, aastor or cod-liver oils. As a confirm- atory test, the skin of the arm may be rubbed with a few drops of the alcoholic solution. s=]uble ilyl &.ao!ute &!coho!, 2nd gives iii alcoholic ao!u:ion & sp&fic co!ofir reaction. After a few hours a vesicular eruption will appear on the part touched.C. A. M. Estimation of Essential Oils in Liqueurs. C. F. Muttelet. ( i l n i z . E'aZsi$c., 1916, 9, 17-22.)-.According to a recent French regulation, alcoholic beverages must not contain more than 0.50 grm. of essential oils per litre. The author's method of estimation is to mix 200 C.C. of the liqueur with 75 C.C. of water, and distil abott 200 C.C.of the mixture. The distillate, containing about 25 per cent. of alcohol, is introduced into a 300 C.C. stoppered flask, which contains 50 grms. of finely powdered recryatallised sodium chloride, and after the addition of 10 C.C. of petroleum spirit the flask is shaken until the salt has dissolved, a few C.C. of water being added if necessary. The shaking is continued f.or about ten minutes, and the flask then allowed to stand. I n the presence of the salt the petroleum spirit is able to extract the essential' oils from their alcoholic solution.The saline layer is extracted twice more with, each time, 5 C.C. of petroleum spirit, and the united extracts dried over anhydrous sodium sulphate, and evaporated by passing a slow current of dry air fhrough the flask, which is meanwhile immersed in water at 30" C.The residue is weighed at intervals of five minutes until two consecutive weighings are identical. I n ;test experiments, the loss of essential oil did not exceed 1 to 2 mgrms. C. A. &I.276 ABSTRACTS OF CHEMICAL PAPERS Estimation of' Essential Oils in Liqueurs. L. Bonnet, ( A m . Falszfic., 1916, 9, 14-16.)-1n testing liqueurs such as anisette and chartreuse 110 C.C.of the sample are mixed with 30 to 40 C.C. of water, and 100 C.C. distilled and mixed with sufficient 95 per cent. alcohol and water to give 150 C.C. of liquid of 50 per cent. alcoholic strength. This is mixed with 40 C.C. of Hiibl's iodine solution and left for three hours in the dark at 15" to 18" C., and the iodine absorption estimated in the usual way.Test experiments with liqueurs containing known quantities of essential oils showed that the factor for calculating the amount of oils in grms. per litre from the iodine value was 1*515, in the case of anisette and 1.498 in the case of chartreuse. I n practice it is advisable to base the coefficient upon the predominating essential oil in the mixture.The blank consists of 150 C.C. of 50 per cent. alcohol. C. A. M. Estimation of the Iodine Value of Essential Oils. R. Marcille. ( A m , FaZsi$c., 1916, 9, 6-11.1-The influence of light upon the estimation of the iodine value of essential oils has already been pointed out (ANALYST, 1915, 40, 152). This influence varies in an irregular manner. I t lowers the iodine value of most essential oils, but in a few cases, notably aniseed oil, it increases the value.I t is advisable t o allow the absorption to proceed for twelve to twenty-four hours, parallel estimations bei- -^I ug ulade in the light and iiz the dark. 3 the icdine :due obtpLined in the 1io.ht; "b-" be greater than that obtained in the dark, it is probable that the mixture contained aniseed or an allied oil.For example, in the dark a sample of aniseed oil gave an iodine value of 1.41 in twenty-four hours, rising to 1-65 after ninety-six hours, whereas in the light it showed 2.03 after thirty minutes, falling to 1-96 after two hours, and 1.79 after Beven hours. Since menthol gives no iodine value, it is possible to dis- tinguish between it and peppermint oil.As a rule, in examining liqueurs it is un- necessary to distil the sample, the presence of sugar not interfering with the estimation of the iodine value. I n the case of certain absinthes, however, differences were observed in the results obtained before and after distillation. * The following are some results recorded by the author. I n each case the iodine solution was left in contact with the oil for three hours in the dark at 25' C.Specific Gravity. 0.991 0.992 1.003 1.027 l*OCiO 1.078 Refractive Index. 1.5609 1.5605 1,5558 1.5518 1.5450 1.5488 1.5480 I Iodine Value. I 1.32 1.38 1.18 0.93 0.43 0.30 0.24 The last sample was a Russian oil which had been kept for six months in a corked flask. Originally its iodine value was 1-24. C. A. M.FOOD AND DRUGS ANALYSIS 27 7 Free Free and j Kitrogen.I Xitrogen. Acid-amidc I Nitrogen. Nitrogen. Estimation of Fat in Cream. L. Lindet. ( A m . Falsijc., 1916, 8, 291-294.)- The author describes a method involving the measurement of the area of the splash of a drop on bibulous paper. w. P. s. Combined Residual acid Nitrogen. ! Non - protein Nitrogenous Constituents of Feeding - Stuffs.H. S. Grindley and H. C. Eekstein. (J. A n z c ~ . Clzem. SOL, 1916, 38, 1425-1431.)-The various forms of nitrogen not precipitated by colloidal ferric hydroxide from a cold aqueous infusion of a feeding-stuff are given in the following table, the figures expressing percentages on the original inaterial : I ... I Go162 0.038 . . . 1 0.262 0.006 ... I 0.082 0.005 ...1 0.132 0.023 ... I 0.285 0.036 Alfalfa hay ... Timothy hay ... Blood meal . . . Maize ... ... Clover hay . . . I I t is apparent that the non-protein nitrogenous constituents consist largely of the forms of nitrogen that result from the decomposition of proteins by hydrolysis, and it is evident that only a small part, if,any, of these constituents can in any way interfere with the application of the Van Slyke method (cf.ANALYST, 1915, 40, 399; 1916, 46) for the estimation of amino acids. w. P. s. Analysis of Spirit of Peppermint. H. L. Thompson. (Amer. J. Pharnz., 1916, 88, 303-308.)-Fifty C.C. of the sample are mixed with water at about 50" C. in a special flask with a graduated neck, and after standing for about four hours the volume of the separated oil is read directly on the scale.An aliquot portion is with- drawn with a pipette, filtered, and weighed, to obtain the weight corresponding to the volume. The weighed quantity of oil is boiled for an hour beneath a reflux con- denser with 25 C.C. of $ alcoholic sodium hydroxide, and the excess of alkali titrated with The number of C.C. used in the saponification multiplied by 9.9 and divided by the weight of oil gives the percentage of menthyl acetate in the oil.The solution is transferred to a separating funnel, and the residual oil (which contains menthol) is washed two or three times and separated. It is then mixed with 1 grm. of anhydrous sodium acetate and 5 C.C. of acetic anhydride, and boiled for an hour on the sand-bath, after which it is washed in a separating funnel, mixed with about 50 C.C.of water, rendered faintly alkaline to phenolphthalein, washed, and separated. A weighed quantity of the acetylated oil is boiled with 25 C.C. of alcoholic sodium hydroxide solution beneath a reflux condenser, and the excess of alkali titrated as before. The number of c.c., multiplied by 7.8 and divided by the weight of oil taken,gives the amount of free and combined menthol in the oil hydrochloric acid.C. A. 31.278 ABSTRACTS OF CHEMICAL PAPERS Detection of Natural and Artificial Pigments in Oleomargarine and Butter. L. S. Palmer and W. E. Thrun. ( J . I d and Eng. Chem., 1916,8,614-615.) -The detection of carotin, the natural yellow pigment of animal fats, in oleomargarine is made necessary by the oleomargarine laws of many States.Cornelison’s test for carotin (J. Anter. Cltem Xoc., 1908, 30, 1475) artificially added to fat is shown to be quoted wrongly in Leach’s ‘‘ Food Inspection and Analysis.” It does detect carotin, as Leach states, but does not distinguish between carotin in genuine butter and carotin prepared from carrots. This accords with Cornelison’s own statement ant1 with the fact now known that the colouring matters are identical.The relation between Martin’s test for artificial pigments in fats and Noore’s test for carotiii artificially added to fats is also confused by Leach (Zoc. cit.). Martin’s general test for artificial colouring matter (ANALYST, 1885, 10, 163 ; 1887, 12, 70) is undoubtedly useful for the preliminary detection of artificial pigments, but is not applicable to the detection of either natural or added carotin in fat.Moore’s test (ANALYST, 1886,11, 163) is not alone specific for carotin added artificially, but is given equally by the carotin of natural fats. Carotin is not dissolved out of the fat in Moore’s test, as he supposed, but merely decolorised by the ferric chloride added. The reaction is shown to be one of oxidation.Although no known modification of Xoore’s test serves to distinguish between natural and added carotin in fats, as a means of detecting carotin it is very useful, and the following simplified procedure is recommended : The rendered melted fat (10 c.c.) is shaken with successive portions of Martin’s reagent (15 parts 95 per cent. alcohol and 2 parts carbon disulphide) until no more colour is extracted.The extracted colouring matter is tested for annatto and other pigments by one or more of the approved methods. All the alcohol is drained off from the fat, and the pigment remaining in the fat is tested for carotin by adding a small crystal of pure ferric chloride to the hot melted fat. After thorough shaking, the fat is extracted with i 0 C.C.of 95 per cent. alcohol. If the pigment was carotin, the fat which separates will, when melted, be seen to be completely decolorised. If only just sufficient ferric chloride be added to oxidise the carotin, the fat will be coloured green by the ferrous chloride, which is a product of the reaction (cf. hgonier-Williams, ANALTST, 1912, 37, 596). G. C. J. Analysis of Saffron. Pierlot.( A m . FaZsi$c., 1916, 9, 24-29.)-Estimation of the amount of ash in saBron ought to be accompanied by a qualitative examination of the ash, since the proportions of ash in pure saffron may vary from 5 to 7-5 per cent. The salts of most frequent occurrence as adulterants are borax, potassium nitrate, sodium and magnesium sulphates, alum, and potassium tartrate.I n one sample barium sulphate was present. The nitrogen varies within much smaller limits. In the case of forty samples of different quality and origin the extreme limits for nitrogen were 2-22 and 2.437 per cent. Apparently, the nitrogen is present in the form of a protein. Pure saffron contains no trace of nitrate, and only a small proportion of ammonia. Fresh samples examined contained 0.015 to 0.04 per cent, of ammonia, while very old samples contained 0.20 to 0.27 per cent.The presence of nitrates may be detected by treating the saffron with sulphuric acid. Pure samples give an indigo- blue coloration, changing rapidly to reddish-brown and then to black, whereas, whenFOOD AND DRUGS ANALYSIS 279 Ash. 0.9( 0.94 1*5C 5.15 4.64 4-72 5-08 4.43 3.95 4-66 4.30 2.13 3.08 2-69 .Q*S8 2.49 2.69 2.26 9-10 1.93 6.35 6.1 7 6.93 7-27 7.01 4.29 11.65 4.60 5-42 7.57 3 *14 3.31 3.74 a nitrate is present, a peach coloration, changing to light rose, is obtained.Nitric nitrogen is readily estimated by reducing the nitrate to ammonia in the usual way. For the estimation of organic nitrogen, the nitric nitrogen must first be eliminated.This may be done by treating 1-5 to 2 grms. of the saffron with 1.5 grms. of pure dry ferrous chloride and 15 C.C. of strong hydrochloric acid. c. A. Rf. Analysis of Spices. C. Arragon. ( A m . Fulsific., 1915, 8, 345-346.)-The following results were obtained on the analysis of various spices, the figures, except i n the case of the water content, expressing grms. per 100 grras.of dry material : _ _ ~ ~- ____ White pepper (Muntok) Black pepper (Batavia) . . . 2 ) 9 2 9 , ... 7 7 , , (Singapore) . . . 9 9 ,, (Java) ... . . . - 7 9 , , (Tellicherri) Long pepper.. . ... Pimento (Jamaica) . . . Cinnamon (Ceylon). . . 9 2 7 , ... ... ,, (China) . . . 9 , 2 7 * ' . Nutmegs (Banda) . . . 3 7 9 9 9 , 7 , 9 9 9 , ... . . . ... ,, (Java) . . . Mace (Banda) . . .9 , Y , Cloves (Zanzibar) . . . Aniseed (Spain) . . . Cumin (Holland) ... Coriander . . . . . . Laurel leaves ... Cardamom (Ceylon) ,, (Malabar) Ginger (Japan) . . . ,, {Bengal) ... Star aniseed ... Aniseed (Japan) ... 9 , 9 , ... Galanga, ... ... ... . . . ... . . . ... ... ... ... . . . ... ... ... ... -.. ... ... ... ... ... ... ... ... ... ... ... ... Watei 10-6 11.4 10.7 10.0 10.2 9.5 7-5 11.3 11-4 9.3 8.2 11.0 10.5 5.8 5.6 3.75 3.0 7.7 1.1.5 5.5 8.5 7.2 9.2 10.5 9.2 4.0 7.8 9.3 ti *5 7.3 3.3 4.4 6*7 1 Esseiitia Fat.* 1 * 8.41 8-45 8.8E 10.59 8.90 8-05 7.4 6-3 1.3 0.9 0.6 1-4 1% L0.2 39.9 39.7 L0.1 36.0 13.7 8.4 8.0 15.5 12-9 I0.6 7.7 2.1 1.8 4.8 3.8 4.9 4.4 5.6 70.0 2-19 3.58 2.13 2.67 2-78 2-47 3.0 4.84 52-74 3.1 2.9 2-9 2.9 6.25 7.32 9.34 6.78 7.45 12.0 13-75 21.1 19.3 7.7 2.3 2.2 4.3 3.3 6.3 1.8 2.9 8.9 2-6 1.7 Reducing 1 Sub- Protein.Cellulose. stances, and Starch. _ _ ! 14-C 15.C 13.0 144 14.2 12.8 12.6 6.9 3.7 4.7 3.7 3.8 3.4 7.8 7.5 7.4 7.1 7-7 6.8 6.9 5.9 5.6 20.2 22.6 14.2 7.0 14.2 11.7 6.3 7.7 4-8 5-8 2.3 6.36 64.2 5.81 ' 65.7 5.54 66.7 11.7 13.0 15.0 7.7 23.0 38.5 35.6 34.9 28.0 28.9 5.66 6.61 6.87 4.72 6.28 4.29 3-30 14.4 15.6 19.2 13.4 24.7 26.6 18.7 13-2 5.0 8.2 29.1 32.1 21.4 46.1 48-6 52.8 51.6 22.5 21.4 21.7 25.1 37.7 35-2 33.0 31.3 29.3 31.9 39.2 40.9 40.3 16.2 16.1 24.6 20.4 21.2 22.3 39.7 58-3 67-1 63.0 13.3 13.0 42 *4 w.P. s. * The essential oil and fat were estimated by weighing the ether extract and removing the former by distillation in steam.280 ABSTRACTS OE CHEMICAL PAPERS Colorimetric Determination of Vanillin in Vanilla. T. von Fellenberg. (Mitt. Lebensmittelzmters. Hyy., 1915, 6, 267-274 ; through J . Chem. SOL, 1916, 110, ii., 355.)-One grm. of the finely divided sample is boiled under a reflus apparatus with four successive quantities of about 20 C.C. of water, the extracts are diluted to 100 c.c., 0.5 grm. of kieselguhr is added, and the mixture filbered. Fifty C.C. of the filtrate are extracted five times with alcohol-free ether, using 150 C.C. of the solvent altogether, the ethereal solution is treated with solid calcium chloride, filtered, evaporated to a small voIurne, and the remainder of the ether removed by a current of air. The residue is warmed to 60” C. with 30 C.C. of water, the solution filtered, and the filtrate diluted to 100 C.C. ; 5 C.C. of this solution are then treated with 5 C.C. of 1 per cent. isobutyl alcohol solution (in 95 per cent. alcohol) and 20 C.C. of concentrated sulphuric acid, and the coloration produced is compared, after forty- five minutes, with that given by a known quantity of vanillin. I t is recommended that the vanillin should be estimated separately in the outer and inner portions of the vanilla pod; in the case of normal vanilla, these two portions contain approxi- mately the ~ a m e quantity of vanillin, and a difference would indicate that some of the vanillin had been extracted from the outer portion.