Page images

obtained remains quite clear and free from morphine crystals. For opium, 3 grams is frequently shaken with a mixture of 15 grams dilute alcohol and 15 grams water and digested during 12 hours. The filtrate is made faintly alkaline with ammonia, and evaporated to half its volume. The solution is made up to its original weight and filtered. 21.25 grams of this filtrate is treated with 5 grams of ether and 0.4 of gram ammonia, and shaken round occasionally during five or six hours. The ethereal layer is taken off with a pipette and passed through two equal filters, on which the morphine is collected and washed twice with 2 c.c. of water each time. After drying at 100°, the morphine is weighed, one paper serving as tare. Of opium extract, 1.5 grams is treated with 10.5 grams of dilute alcohol, and 10.5 grams of water without heat, and filtered. The weighed filtrate rendered slightly alkaline by ammonia is boiled down to one-half, made up to its original weight with water, and filtered. 15 grams of the filtrate is treated with ether and ammonia as above. Of Tinctura opii simplex or crocata, 25 grams is taken, made slightly alkaline with ammonia, and treated as above. J. T.

Estimation of Cinchonidine in Quinine Sulphate. By L. SCHÄFER (Arch. Pharm. [3], 25, 64-72).-After reviewing several methods, the author gives one for estimating 1 per cent. or less of cinchonidine in quinine sulphate, based on the extremely slight solubility of quinine oxalate in water in presence of a small excess of potassium oxalate, and the relatively easy solubility of cinchonidine oxalate in such a solution. 2 grams of quinine sulphate is dissolved in a small tared flask in 55 c.c. of boiling water, and 0.5 gram of neutral, crystallised potassium oxalate in 5 c.c. of water is added. The liquid is made up to 625 grams and cooled for half an hour in water at 20°, with occasional shaking to and fro, and then filtered. If on the addition of one drop of officinal aqueous soda to the filtrate no turbidity appears, the quinine sulphate contains less than 1 per cent. of cinchonidine sulphate. In the presence of 1 per cent. of the latter salt, a turbidity or a precipitate of cinchonidine appears. Quantitatively, 5 grams of quinine sulphate is taken, and an aliquot part of the filtrate is treated with aqueous soda; the cinchonidine is collected. Since a certain amount of cinchonidine remains in solution, and a little also goes down with the quinine oxalate, it is necessary to apply a slight correction to the amount found. Numerous experiments show that this correction should be 0.04 gram cinchonidine for each 100 c.c. of solution originally taken. Further, if more than 4 per cent. of cinchonidine is present, a more dilute solution should be employed, as the process is expressly intended for small quantities only. The test also indicates small quantities of quinidine and cinchonine sulphate when present; indeed, the conditions are more favourable in the case of these compounds, as they are not carried down by the oxalate precipitate.

J. T.

Testing Quinine Sulphate. By O. SCHLICKUM (Arch. Pharm. [3], 25, 128-129).-Employing De Vrij's chromate method (this vol., p. 404), the author finds that not only quinine but also cinchonine

forms a chromate soluble in 2000 parts of water at moderate temperatares, whilst quinidine and cinchonidine chromates are much more soluble in water. On precipitating a quinine solution by means of normal potassium chromate, and allowing it to remain four or more hours, the filtrate remains unchanged on the addition of soda if the quinine salt is pure. If the quinine salt contains cinchonine, quinidine, or cinchonidine in not too minute traces, the soda produces a turbidity either at once or after some time. The method detects cinchonine sulphate to per cent., and cinchonidine or quinidine sulphate to 1 per cent. In testing other neutral quinine salts it is not necessary to convert them into sulphate. Acid quinine salts require conversion into neutral ones, say by evaporation to dryness with ammonia.

J. T.

Colour Reactions of Picric Acid and Dinitrocresol (Victoriayellow). By H. FLECK (Chem. Centr., 1887, 99).-When solutions of these substances are evaporated in a porcelain dish, and the residue moistened with a little 10 per cent. hydrochloric acid, a small piece of pure zinc added, and the dish allowed to remain for some hours without warming, a fine blue colour is developed in the case of picric acid and a bright blood-red with dinitrocresol. These reactions are useful for examining artificially-coloured farinaceous foods, the alcoholic extract of which should be employed. G. H. M.

Determination of Tannin in Sumach. By J. MACAGNO (Chem. Centr., 1887, 125).-The author has compared Löwenthal's method for the determination of tannin with those of Davy and Gerland. He finds that Davy's method, which consists in precipitating the tannin with gelatin, drying, and weighing the precipitate, and multiplying the weight by the factor 0'4, gives results both with pure tannin and also with sumach which stand in the ratio to results obtained by Löwenthal's method as 53:34: 100; whilst Gerland's method (precipitation of the tannin with tartar emetic solution in the presence of ammonium chloride; the reagent is prepared by dissolving 2·611 grams dry tartar emetic in a litre of water, 1 c.c. equals 0005 gram tannin) gives results which, when compared with Löwenthal's method, stand in the ratio of 2: 3.

G. H. M.


General and Physical Chemistry.

Absorption-spectrum of Liquid Oxygen and of Atmospheric Air. By K. OLSZEWSKI (Monatsh. Chem., 8, 73-77).-Liquid oxygen, examined with a thickness of 12 mm. and at a temperature of —181·4°, gives absorption-bands, the middle of which correspond with the wavelengths 628, 577, 535, and 480μ; the band 628 is characterised by its width (634-622), and the band 577 by its intensity; the bands 535 and 480 do not appear to be present in the solar spectrum.

Liquid air was examined at -191°, and with a thickness of 12 mm. No other absorption-bands besides those of oxygen were observed, but bands 628 and 577 were not so strong as with pure oxygen.

G. H. M.

Red Fluorescence of Alumina. By L. DE BOISBAUDRAN (Compt. rend., 104, 824-826).-The author has previously found (this vol., p. 538) that alumina prepared from pure aluminium chloride shows no phosphorescence even in the phosphoroscope. An aqueous solution of the chloride was left exposed to the air in a glass vessel for several days, and was then evaporated to dryness, and the residue strongly heated. The alumina thus obtained gave no fluorescence in a vacuum, and only a very feeble tint in the phosphoroscope.

Alumina from aluminium chloride, which has been very strongly heated and shows no phosphorescence in the phosphoroscope, shows a brilliant red phosphorescence when mixed with a very small quantity of chromium.

In order to ascertain whether the fluorescence shown by alumina from alum is due to the presence of impurities, ammonia alum was recrystallised seven times from a slightly acid solution. Alumina prepared from the seventh crystallisation, and very strongly heated, gave no red fluorescence in a vacuum, but a moderately intense violet fluorescence, becoming indigo with a weaker current. In the phosphoroscope, it showed a very feeble greenish phosphorescence with a tendency to become red at some points. Alumina from the fifth crystallisation gave no red in a vacuum, but a somewhat marked pale green fluorescence, becoming violet with a weaker current. The alumina from the third crystallisation gave the red fluorescence, and the mother-liquor gave the spark spectrum of chromium.

Alumina precipitated from the seventh crystallisation by ammonia, and very strongly heated, showed no trace of the red fluorescence in a vacuum, but gave a mixture of a feeble green and a still feebler violet fluorescence. If this alumina is mixed with 0.0000186 of its weight of chromium oxide, it shows a beautiful rose-red fluorescence. The author considers that these facts, together with those previously described, show that the red fluorescence is really due to the presence of minute quantities of chromium. C. H. B.


2 t

Molecular Refraction of Carbon Compounds of High Dis. persive Power. By R. NASINI (Gazzetta, 17, 48-55, and 55—64). In these papers the author criticises Brühl's conclusions regarding the relations between molecular refraction and chemical constitution. The papers are mainly controversial, and contain neither fresh determinations nor conclusions. V. H. V.

Formation of the Electric Arc without Contact of the Electrodes. By G. MANEUVRIER (Compt. rend., 104, 967-969).— The electrodes are enclosed in an air-tight glass vessel provided with a three-way stopcock, the electrodes being connected with a source of alternating currents by means of platinum wires fused into the glass. The apparatus is attached to an air-pump, and the pressure inside reduced until a violet silent discharge takes place between the poles. The stopcock is then turned so as to admit a small quantity of air, and under the influence of the sudden increase of pressure, the silent discharge between the poles is transformed into an arc. When the arc has formed, the stopcock is closed, and in this way an arc is obtained in what is practically a vacuum, and thus many of the causes which interfere with the constancy of the arc are eliminated; there is, for example, no combustion of the carbons. The silent discharge passes when the pressure in the globe is 5-6 mm., and the arc is formed at pressures between 30 and 150 mm.

C. H. B.

Thermic Expansion of Liquids at Various Pressures. By G. P. GRIMALDI (Gazzetta, 17, 18-31).-In continuation of experiments on the expansion of liquids at various pressures (Abstr., 1886, 498), the author gives determinations for chloroform at temperatures varying from 0-80° and pressures of 1 to 15.5 metres, and of pentane between 1 and 100°, and pressures of 1 to 22 metres. The equation expressing the expansion in terms of temperature is of the form ▲ = at + bť2 + ct3, and the values for the constants a, b, c are given for chloroform and pentane at pressures of 1 and 15.5 metres and 12 and 22 metres respectively.

TaV K = B


[ocr errors]

The various formulæ, expressing the dilatation in terms both of temperature and pressure, which have been proposed by Duprè, Heen, and van der Waals are fully discussed. The formula most in accordance with the results is a modification of Dupré's, and is expressed thus: in which K = aA2 when t = 0°, T is the absolute temperature, & the time coefficient of dilatation at pressure p, ẞ the coefficient of compressibility at T, and K a constant dependent on the nature of the liquid examined. The difference between the observed and calculated values is most marked at the extreme limits of temperature, at which the determinations are least exact, and it is noted that a small error in the experimental value for the expansion influences to a considerable degree the value for the constants a and B. The coefficients calculated according to Heen's equation give results lower than those observed, and these differences rapidly increase with increase of temperature. V. H. V.

Latent Heat of Vaporisation of Certain Volatile Substances. By J. CHAPPUIS (Compt. rend., 104, 897-900).-The apparatus consists of a cylindrical glass reservoir closed at the bottom, and containing the liquid to be evaporated. This receiver terminates in a serpentine capillary tube united to an ordinary delivery tube, and to the free end of this a steel stopcock with a lateral tubulus is cemented.

The receiver containing the liquid under examination is weighed and placed in a Bunsen's ice-calorimeter, in which both it and the serpentine tube are completely surrounded by mercury. The stopcock is then opened, and the vapour is allowed to escape very slowly so that the reduction of pressure which is essential to vaporisation may be kept as small as possible. After the usual readings have been made, the apparatus is again weighed. The loss of weight gives the difference between the weight of the liquid which has been volatilised and the weight of its saturated vapour which occupies the same volume; from this, the weight of the liquid evaporated is readily calculated. The following results were obtained :

Methyl chloride.....
Sulphurous anhydride




Further experiments, which will be described in a subsequent paper, show that the rate of vaporisation exerts considerable influence on the results, but if it does not exceed 8 to 16 mgrms. per minute, the latent heat of vaporisation is constant. Within the same limits, the temperature of the apparatus in which evaporation takes place is not reduced below 0.3°. C. H. B.

The Calorimetric Bomb. By BERTHELOT and RECOURA (Compt. rend., 104, 875-880).-The calorimetric bomb (Abstr., 1886, 756) consists of three metals, platinum (interior), steel, and brass (stopcock). Its water value may be calculated from the weight of these metals and their specific heats, or may be directly determined by one of three methods, namely: (1) by burning in the bomb in the calorimeter two different weights of the same substance, one being twice or three times as great as the other; the thermometric measurements give equations which contain the value of the unknown quantity; (2) by introducing into the water of the calorimeter containing the bomb a known quantity of water at a definite temperature; (3) by introducing into the water of the calorimeter a known weight of concentrated sulphuric acid, a previous experiment being made with the same quantity of water and sulphuric acid, but without the bomb in the calorimeter. Direct determination gave 343·9 grams; whilst the calculated water value was 3447 grams.

Ten minutes after compressing the gas, or allowing it to escape, the rate of cooling of the apparatus resumes its normal value. The compressed oxygen from the pump is passed through a copper tube heated to redness in order to oxidise any organic matter, and is cooled to the ordinary temperature before entering the bomb.

Three of these bombs are in existence, and a comparison of the heat

« PreviousContinue »