Page images

potassium dichromate, and treats the mixture with sulphuric acid in a Will's carbonic acid apparatus in the ordinary way for a carbonic acid determination. Gentle boiling is required to complete the oxidation of the glycerol to carbonic anhydride and water. In analysing sweet wines, the glucose they contain may be destroyed by evaporation to dryness with baryta-water before extracting the glycerol with alcohol. Cane-sugar must first be inverted by boiling with hydrochloric acid. The crude glycerol contains impurities which yield carbonic anhydride on oxidation. The amount of these precipitated by ammoniacal solution of lead acetate is said to be constant and to be equivalent to 0.035 gram of glycerol per 100 c.c. of wine. Further experiments are promised.

M. J. S.

Estimation of Glycerol and its Non-volatility with Aqueous Vapour. By O. HEHNER (Analyst, 12, 44-46, and 65-67).—The glycerol is oxidised as in Legler's process (preceding Abstract) by an excess of potassium dichromate and sulphuric acid, but the author measures the amount of dichromate reduced. One part of glycerol requires 7-486 parts of dichromate. The oxidising solution contains 80 grams of potassium dichromate, and 150 c.c. of strong sulphuric acid per litre. The unreduced excess is determined by titration with iron and standard dichromate. In moderately strong solutions (12 grams per litre), two hours' heating is sufficient to ensure the complete oxidation of the glycerol, but in highly dilute solutions a further addition of sulphuric acid is requisite. Such dilute solutions can, however, be concentrated by vigorous boiling without the loss of a trace of glycerol. Even a 50 per cent. solution boiled for two hours in a covered beaker, with continual replacement of the evaporated water (257 c.c.), did not suffer an appreciable loss of glycerol, whilst the loss from a 74 per cent. solution was very insignificant.

M. J. S.

Determination of Moisture in Starch. By F. W. DAFERT (Chem. Centr., 1887, 567).—Experiments are described to ascertain the temperature at which starch suffers the maximum loss in weight without appreciable decomposition. At 120° the loss was 11:31 per cent.; at 105-107°, 10.89 per cent.; and at 100° in a vacuum 11.9 per cent.; in the first case, the starch was slightly changed, but in the last it was unaltered. It is recommended that determinations of moisture should be conducted by this method. V. H. V.

Logwood Test for Alum in Bread. By W. C. YOUNG (Analyst, 12, 29-32, and 145-147).-Sour bread gives the same blue coloration with logwood as bread to which alum has been added. Specimens of bread which gave no indication with the logwood test, all showed the blue colour intensely after being moistened with very dilute acetic acid. Since aluminium phosphate is perceptibly soluble in cold, though not in hot, acetic acid, it is possible that this may account for the reaction after acidifying. M. J. S.

Estimation of Hydrocyanic Acid. By O. LINDE (Arch. Pharm. [3], 25, 690).—In the determination of hydrocyanic acid in bitter

almond water according to the Paris Pharmacopoeia, the author proceeds as follows:-The bitter almond water is diluted with two volumes of water as the reactions then proceed most rapidly; basic magnesium acetate is added at once in considerable excess of the potassium chromate solution, not "some" drops, but only two drops per 27 grams of water to be added; of the silver solution almost sufficient to precipitate the whole of the acid should be added at once; the titration is best made by daylight, as the persistent red colour is not so well seen by gas- or lamp-light. J. T.

Estimation of Thiocyanic Acid. By P. KLASON (J. pr. Chem. [2], 36, 74—77).—If hydrochloric acid is not present, thiocyanic acid may be accurately determined by titration with silver nitrate. If hydrochloric acid is present the determination must be made by oxidation with permanganate. Here, however, the degree of concentration influences the result-which is always too low. If the concentration is not less than decinormal, the result is 1 per cent. too low, and this should be added to the amount found. L. T. T.

Estimation of Carbonic Acid in Beer. By C. A. CRAMPTON and T. C. TRESCOT (Amer. Chem. J., 9, 290–293).-The cork of the bottle is pierced by a champagne tap and the gas passed through an Erlenmeyer's flask (to retain any foam), then dried by sulphuric acid and calcium chloride, and absorbed by soda-lime; the sample is afterwards heated to 80°, and the gas swept into the soda-lime tube by a current of pure air. The authors find an average of almost 0.4 per cent., most authorities giving an average of 01 to 02 per cent.; a higher amount of carbonic acid indicates either the addition of sodium bicarbonate or a state of after-fermentation. H. B.

Estimation of the Free Acid in Tannin Liquor by Titration. By R. KоCH (Dingl. polyt. J., 265, 33-41). In a previous communication (this vol., p. 871), the author recommended the use of albumin for the separation of tannin and the decolorisation of tannin liquor. He now finds that a solution of gelatin is preferable, as the precipitation of albumin by the aid of heat and the subsequent cooling is found to be inconvenient. It is necessary, however, to adjust the concentration of the gelatin solution to the amount of tannin in the liquor, otherwise difficulties in connection with the filtration and washing of the precipitate will arise. A solution containing 2 grams of gelatin in a litre of water is applicable to most liquors. The process is described in detail in the paper, and several test analyses are given. The author also states that powdered skin cannot be used for decolorising tannin liquor for the purpose of determining the free acid contained therein, owing to the fact that it absorbs acid from such liquors.

D. B.

The Lactocrite; a New Apparatus for Determining Fat in Milk. By H. FABER (Analyst, 12, 6-11).-The principle of this apparatus and method introduced by De Laval, consists in the separation of the fat by centrifugal force, after the caseïn has been

brought into such a condition that it does not prevent the union of the fat globules. This is effected by heating the milk with an equal volume of glacial acetic acid containing 5 per cent. by volume of concentrated sulphuric acid. The mixture is then introduced into a graduated tube, and several of these tubes at once are rotated for a few minutes in a specially constructed disc. The column of butterfat obtained gives the percentage by a simple reading on the graduated tube. The results can be trusted to within 0.1 per cent. in whole milk, but in skimmed milk are about 0.2 per cent. too low.

M. J. S.

Reichert's Distillation Process. By A. H. ALLEN (Analyst, 12, 11-13). Porpoise oil and some samples of whale oil give nearly as large a percentage of volatile fatty acids as butter does. The appearance of solid fatty acids in the distillate is an indication of the presence of cocoanut oil. They should be filtered off before titrating. It is desirable to saponify the fat in a flask or closed bottle, since saponification in an open basin is found to give low results. The form of the distillation apparatus is of little importance.

M. J. S.

Volumetric Estimation of Urea. By G. CAMPANI (Gazzetta, 17, 137-141).—The method described is based on the reaction of nitrous acid on urea, and estimation by alkalimetry of the carbonic anhydride evolved. In a tube of 150 to 200 c.c. capacity 20 c.c. of a 10 per cent. solution of potassium nitrite is introduced, then 2 c.c. of urine, and lastly 2 c.c. of 5 per cent. sulphuric acid. The tube is then fitted with a delivery tube, the contents gently warmed and the carbonic acid evolved is collected in lime-water of known strength. When the reaction is complete, the amount of lime neutralised by the carbonic anhydride is determined in the usual manner by a standard solution of oxalic acid; 1 mol. of carbonic anhydride corresponds with 1 mol. of urea, thus: CO(NH2)2 + 2HNO2 = 3H2O + 2N2 + CO2. The results obtained in test analyses are fairly concordant with one another and with those obtained by Liebig's method. It is necessary that the reaction should be allowed to proceed slowly; otherwise nitrous acid is evolved. V. H. V.

Estimation of Quinine Sulphate. By O. HESSE (J. Pharm. [5], 16, 213–215).-1 gram of the sulphate dried at 100° is agitated with 20 c.c. of water at 60° and filtered after cooling; 5 grams of the filtrate is placed in a somewhat narrow test-tube; 2 c.c. of ether and five drops of ammonia are added; the tube is now stoppered and carefully agitated. The ethereal solution ought to be limpid, and should not deposit crystals. The author thinks that Paul's method gives good results when modified as follows: 5 grams of sulphate is dissolved in boiling water and allowed to crystallise; the motherliquor is set aside and the crystals are dissolved in 120 c.c. of boiling water, recrystallised, and the process is repeated a third time, &c., the mother-liquor always being set apart. Three crystallisations suffice


4 h

with a salt containing 5 per cent. of cinchonidine sulphate; five crystallisations are indispensable when 9 per cent. is present. The mother-liquors of the three first crystallisations are evaporated to dryness, the residue is dissolved in dilute sulphuric acid, water is added to 20 c.c., and this is agitated with 16 c.c. ether and excess of ammonia. After 24 hours, the crystals of cinchonidine are collected. Similarly the remaining mother-liquors are made into a volume of 8 c.c., and treated with ammonia and 2 or 3 c.c. of ether; the crystals obtained are weighed with the preceding crop. The results thus obtained agree sensibly with those afforded by de Vrij's bisulphate method. J. T.

Testing Quinine Sulphate. By G. KERNER and A. WELLER (Arch. Pharm. [3], 25, 712-738, and 749-765).-The ammonia test for commercial quinine sulphate was proposed by one of the authors at a time when South America was the chief source of the salt, and the compound was almost free from cinchonidine; more recently Asiatic products have appeared in the market, in which cinchonidine often occurs in very considerable proportions. To meet the new conditions, the following method is proposed :-The sample of quinine sulphate is maintained at 40-50° until completely disintegrated, 2 grams in 20 grams of water is kept at 60-65° for half an hour with frequent stirring. The temperature is now brought to 15° and maintained at that for two hours with stirring, care being taken to have the temperature 15° exactly before the subsequent filtering. 5 c.c. of the filtrate is treated with ammonia of 0.96 sp. gr. until the precipitated quinine is again exactly dissolved. The amount of ammonia required should not exceed 10 c.c. The authors conclude that none of the methods hitherto proposed for the exact separation of cinchonidine and quinine are really exact. The same remark holds for the estimation of hydroquinine.

J. T.

Detection of Hop-substitutes in Beer. By A. H. ALLEN (Analyst, 12, 107-114).-A litre of the beer is evaporated to about 300 c.c. and precipitated hot with normal lead acetate, which is preferable to the basic or ammoniacal acetate of older processes. This throws down lupulin and hop-resin completely, leaving all or nearly all hop-substitutes in solution. After filtration, first hot and then cold, the lead is removed by hydrogen sulphide and the filtrate further concentrated. It is then acidified with sulphuric acid and shaken repeatedly with chloroform, which extracts absinthin, anthemin (chamomiles), quassiin, calumbin (colchicine, colocynthine, berberine, gentipicrin, picric acid, and picrotoxin imperfectly); ether is next employed to extract chiratin (chiretta), colocynthin, gentipicrin, picric acid, and picrotoxin. A further treatment with etherchlorof orm after addition of ammonia completes the extraction of berberine and colchicine. If the residue from one of these solvents has a bitter taste, the presence of some hop-substitute is certain.

Quassiin, prepared by the above process, does not reduce Fehling's

solution or ammoniacal silver nitrate, and gives no coloration with strong sulphuric acid, or with nitric acid of 1.25 sp. gr., even on warming. It gives a mahogany-brown colour with ferric chloride. Its chloroform solution treated with an excess of bromine and then with ammonia gives a bright yellow colour (Christensen). A solution of 1 part of quassiin in 500,000 of water has a strong and persistent bitter taste. Gentipicrin and menyanthin reduce ammoniacal silver nitrate. With sulphuric acid, the former gives a red colour, the latter a yellowish-brown becoming violet-red on warming.

The chloroform extracts from chamomiles, calumba, colocynth, cocculus, and chiretta give no reaction with bromine and ammonia. The ether residue from chiretta, gives a straw-yellow colour changing to dull purplish-brown. Chiratin does not reduce Fehling's solution; it gives a copious precipitate with tannin. M. J. S.

Estimation of Ipecacuanha. By F. RANSOM (Pharm. J. Trans. [3], 18, 241-242). The powdered ipecacuanha is thoroughly exhausted with chloroform that has been rendered alkaline by previous agitation with a strong solution of ammonia. An extraction apparatus is employed by means of which the exhaustion can be completed with hot chloroform. The solution is then agitated with dilute sulphuric acid, and the emetine in the acid solution estimated volumetrically by Mayer's reagent.

R. R.

Detection of Aniline Colours in Wine, &c. By C. O. CURTMAN (Zeit. anal. Chem., 26, 555-556).—The formation of isonitriles when aniline-derivatives are warmed with potash and chloroform serves for the direct detection in wine of even minute traces of many of the aniline colours. The evolution of the isonitrile is accelerated by adding excess of strong sulphuric acid. M. J. S.

Estimation of Indigo. By G. MANNLEY (Chem. Centr., 1887, 605): -In order to determine the proportion of indigo in a sample, the moisture is first estimated, and the dried mass introduced into a testtube moistened with alcohol and a fine emulsion of grape-sugar in alcohol, and a few c.c. of soda solution added to it. The tube is then filled with alcohol, agitated for some time, left for 24 hours, and then poured into dilute sulphuric acid and oxidised by agitation. Hydrogen peroxide can also be used, and is to be preferred, because the precipitate in this case does not adhere so much to the walls of the vessel. The whole is allowed to remain for two days, filtered through a tared filter, dried and weighed. A gram of substance is a convenient quantity for the determination. V. H. V.

Testing Indigo Dyes on Fabrics. By W. LENZ (Zeit. anal. Chem., 26, 535-555).—In the course of an enquiry into the genuineness of the indigo dye on a certain fabric, an extensive examination of the methods of testing blue dyes was made, and those at present in use were found not to be altogether trustworthy. Stuffs were

« PreviousContinue »