Page images

connecting the measuring tube with a closed volume of moist air. The eudiometer is connected at the bottom with a movable mercury receiver, and at the upper end terminates in a Greiner-Friedrich slanting-bore three-way tap with two capillary outlet tubes. One of these tubes serves to draw in the gas to be analysed, the other is connected by a rubber joint with a small U-shaped mercury manometer, the other end of which is similarly connected with the air-chamber. The eudiometer and air-tubes are placed side by side in a cylinder of water. L. T. T.

Estimation of Phosphoric Acid. By MALOT (J. Pharm. [5], 16, 157-159). This communication is supplementary to one which appeared in the July number of J. Pharm. on the titration of phosphoric acid by means of uranium nitrate. The green lake formed by uranium oxide with the colouring matter of cochineal being decomposed by dilute acids, not excepting acetic acid, it is necessary to operate with a liquid slightly acidified with acetic acid. Hence, instead of adding 5 c.c. of sodium acetate solution to the assay slightly acidified with nitric acid, it is better after having brought the colour of cochineal to violet by means of ammonia to remove this colour by adding one or two drops of acetic acid. This also produces the best conditions for the precipitation of phosphoric acid by means of uranium acetate. The solution should be kept boiling, especially at the termination of the operation, which is reached when a definite green tint is produced, neither modified by boiling nor by the further addition of uranium salt. Instead of uranium nitrate, it seems preferable to employ the acetate, whose solution better preserves a constancy of composition if care has been taken to boil during a few hours in order to precipitate phosphate, which is always present in commercial acetate. Alcoholic tincture of

cochineal serves as well as the solution obtained by boiling in water. The ammonium magnesium phosphate precipitate need not be dissolved off the filter, but may be thrown along with the latter into a 200 c.c. flask for the titration. This precipitate may be impure, either from the presence of citric or molybdic acid. Either of these acids interferes with the accuracy of the assay. It is convenient to dissolve the ammonium phosphomolybdate in concentrated ammonia, and to precipitate with citro-magnesium solution in a liquid strongly ammoniacal, and to wash once at least with strong ammonia, finishing the washing with dilute ammonia to remove the last traces of citric acid. J. T.

Examination of Crude Soda Lyes and Red Liquors. By W. KALMANN and J. SPÜLLER (Dingl. polyt. J., 264, 456–459).—In examining crude soda liquors, it is the practice to calculate the total consumption of iodine as sodium sulphide, the separate estimation of sodium sulphite and sodium thiosulphate being rarely attempted owing to the want of a rapid and accurate method. The authors recommend a process based on the insolubility of barium sulphite and the solubility of barium thiosulphate in alkaline solutions. estimation is performed in the following manner :-(1.) The total


alkalinity is determined in a measured volume of the liquor under examination by titration with normal acid, methyl-orange being used as indicator. The acid consumed equals sodium carbonate + sodium sulphide, sodium hydroxide, one-half sodium sulphite (Na2SO, is alkaline and NaHSO, neutral to methyl-crange). (2.) An equal volume of the liquor is titrated with a decinormal solution of iodine, the volume consumed corresponding with the sodium sulphide + the sodium sulphite, the sodium thiosulphate. (3.) Twice the volume of liquor as that used in (1) and (2) is precipitated with an alkaline zinc solution, and the mixture made up to a certain measure, one-half of which is filtered, acidified, and titrated with decinormal iodine solution. The iodine consumed equals sodium sulphite + sodium thiosulphate. (4.) Three or four times the volume of liquor used in (1) and (2) is treated with an excess of a solution of barium chloride, the mixture made up to a known volume with water, and filtered. (a.) One-third or one-fourth (as the case may be) of the filtrate is titrated with normal acid, the amount used corresponding with the sodium hydroxide + the sodium sulphite. (b.) A new third or fourth of the filtrate is acidified and titrated with decinormal iodine solution, the iodine consumed being equal to sodium sulphite + sodium thiosulphate. The calculation is made as follows:

2 .4b





[blocks in formation]




= B c.c. decinormal iodine solution
corresponding to......
= C c.c. decinormal iodine solution

[blocks in formation]


corresponding to...... Na2S2O3. =D c.c. normal acid corresponding


(4a+A) = E c.c. normal acid corresponding


The results of several test analyses are given.



D. B.

Estimation of Titanic Acid. By L. LEVY (J. Pharm. [5], 16, 56-61).—Titanic acid is estimated by fusing the material with hydrogen potassium sulphate, extracting with cold water the mixture of titanate and sulphates thus formed, boiling for six hours to precipitate the titanic acid, calcining, and weighing. In presence of alkalis, magnesium, zinc, aluminium, or copper oxides, it is only necessary to fuse with the sulphate as above, dissolve the mass in water acidified with a sufficient quantity of sulphuric acid, neutralise with potash or ammonia, then add 0-5 gram of sulphuric acid to each 100 c.c. of solution, and finally boil for six hours, constantly replacing the water driven off. This method is not applicable in the case of iron sesquioxide. The results are high. J. T.


General and Physical Chemistry.


Oxygen in the Sun. By J. TROWBRIDGE and C. C. HUTCHINS (Phil. Mag. [5], 24, 302-310).-In these experiments, a powerful alternating current was caused to pass between electrodes of aluminium, and the spectrum, obtained by a grating, photographed on one half of a photographic plate, then, without altering the arrangement of the apparatus, sunlight was admitted, and its spectrum photographed on the other half of the plate. The wave-lengths of the air or sun spectra were tabulated. The authors point out that in order to be certain of the existence of an element in the sun, the coincidence of a large number of the lines of the element in position and grouping with the dark lines of the solar spectrum is necessary, or else a general similarity in the character of the lines; they find no such coincidence for, 1 oxygen so far as they have examined (wave-lengths 3749-8-5033·85). They find that the bright lines of Draper's spectrum vanish in their high-dispersion apparatus, and contain numerous dark lines, and, moreover, that there is no general coincidence between the oxygen lines and the bright spaces of the solar spectrum. H. K. T.

Existence of Carbon in the Sun. By J. TROWBRIDGE and C. C. HUTCHINS (Phil. Mag. [5], 24, 310–313).—The authors find that the solar spectrum near H contains dark lines exactly agreeing with the spaces between the bright lines of the flutings of the carbon spectrum. The carbon spectrum is wanting in the green and blue. This may be due to the effect of vapour in the sun's atmosphere, the lines due to any element being unaltered, obliterated, or reversed, according to the temperature of the vapour through which the light passes. The temperature of the luminous vapour itself may also vary. Further the spectrum from one element may be partially obliterated by the passage of the light through the vapour of another element. The photographic spectrum of carbon can be obliterated in the green and blue by superimposing upon it that of iron, of nickel, and of cerium. The fluted carbon spectrum of the voltaic arc is due to a reversal of the continuous spectrum of the ignited carbon by its own vapour; hence the temperature of the sun's atmosphere where the carbon is volatilised must approximate to that of the voltaic arc.

H. K. T. Existence of Certain Elements and Discovery of Platinum in the Sun. By C. C. HUTCHINS and E. L. HOLDEN (Phil. Mag. [5], 24, 325–330).—The spectra of different metals in the voltaic arc are compared with the solar spectrum by means of apparatus already described (see preceding Abstracts). The authors confirm the results of Lockyer for the metals cadmium, bismuth, silver, potassium, and lithium, whilst they find insufficient coincidence for the metals lead, cerium, molybdenum, uranium, and vanadium. Platinum, which has not hitherto been found in the solar spectrum, gives 64 lines between


4 c

wave-lengths 4250 and 4950, 16 of which agree with solar lines, whilst seven are doubtful. H. K. T.

Spectrum Analysis. By A. F. SUNDELL (Phil. Mag. [5], 24, 98106). The author examines the spectra of considerable thicknesses of rarefied gases at low temperatures. The gases were contained in glass tubes, viewed end on, the ends of the tubes being coated with tinfoil connected with the conductors of a Holtz machine.

Air first became luminous at a pressure of 10 to 12 mm.; at 8 mm. the light took up peculiar stratifications, which oscillated rapidly. The spectroscope showed a large number of bands, which are tabulated. With pressures less than 0.0003 mm., the luminosity ceased. The author finds that even with a tube which had been frequently exhausted during a period of four months, evidence of air appears shortly after exhaustion. If the discharge is made to take the form of sparks, luminosity ceases at a pressure of 0.004 mm. Hydrogen, oxygen, and nitrogen were also examined. The first was luminous in the highest vacua; oxygen was brightest at 0.2 mm.; nitrogen gave the same spectrum as air. H. K. T.

Radiant Matter Spectroscopy: Examination of the Residual Glow. By W. CROOKES (Proc. Roy. Soc., 42, 111-131).—The author has devised the following modification of Becquerel's phosphoroscope. The phosphorescent substance in an exhausted bulb is excited in the usual way by the spark from the secondary of an induction coil, and is viewed through a rotating disc having 12 symmetrically placed apertures near its margin. A brass cylinder, one end of which is cut into 12 teeth, is fixed upon the axis of the disc, and two springs, one pressing on the continuous part of the cylinder, and the other (adjustable) rubbing over the teeth, serve to complete the circuit through a battery and the primary of the coil. By adjusting the movable spring the spark from the secondary can be made to coincide with the passage of an aperture between the substance under examination and the eye or a spectroscope, or to precede it by a short interval of time, easily calculated when the velocity of the disc is known. By moving the adjustable spring towards or from the bases of the teeth, the relative duration of the makes and breaks can be varied at pleasure. Since the line spectrum of the residual gas in the bulb has no appreciable duration, much lower vacua are necessary for the examination of the residual glow by this method.

When yttria, free from samaria, is thus examined, it is found that all the bands in its phosphorescent spectrum do not appear at the same speed of rotation. As this is gradually increased they come into view in the order Gß (545), greenish-blue, and Ga (482), deep blue (duration 0.0035 sec.); Gd (574), citron (0.0032 sec.); Gy (647), deep red (0.00175 sec.). At an interval of 0.00125 sec. Go and GB are equally bright, and Gn just visible; and at 0.000875 sec. all are seen of their usual brightness. The author confirms the absence of the citron band from the spectrum of phosphorescent yttric phosphate.

Elsewhere the author has pointed out that 1 per cent. of lime added to a badly phosphorescing substance containing yttrium or samarium

brings out the phosphorescence well. In the former case, it diminishes the sharpness of the line Go, but increases its brightness, as well as that of Ga. In the phosphoroscope, lime is observed also to change the order in which the yttric lines appear to Go, Ga, Gß, Gŋ.

The author attributes this action to the long residual phosphorescence of the lime, which induces or assists certain yttric vibrations (G), and is antagonistic to others (GB). The influence of baryta and strontia on the phosphorescence of yttria has also been studied, and the author gives the order of appearance and intensities of the lines. The results cannot well be condensed, the action of each alkaline earth not only being fundamentally different, but varying with the relative proportions of the mixtures. An interesting result is obtained on adding strontia to a mixture of yttria and samaria and viewing it in the phosphoroscope with the wheel rotating rapidly. The line Go is then completely suppressed, and a spectrum is obtained undistinguishable from that of Marignac's Ya (Proc. Roy. Soc., 40,


Many earths and oxides, or mixtures of these with lime, have been examined, mostly as sulphates. The results are given in more or less detail.

Arsenious acid, cadmium, copper, lead, and nickel sulphates, tin, tungsten, and uranium, in small quantity with lime, give the phosphorescence of the latter.

Alumina, giving the crimson line, has a very persistent residual glow. In the phosphoroscope, rubies shine with great brilliancy.

Antimony oxide (5 per cent.) with lime phosphoresces white, with a broad space in the yellow. In the phosphoroscope, the glow is green and very strong: the red and orange are obliterated.

Barium (5 per cent.) with lime, bright reddish-orange phosphorescence, with specks of yellow and violet. The spectrum is continuous, faint in the red, strong in the green, and with a broad black band in the orange.

Bismuth (15 per cent.) with lime, bright reddish-orange. The spectrum shows a broad dark band in the red and orange, a black band in the yellow, and great concentration in the green and blue. In the phosphoroscope, red and orange are obliterated.

Calcium sulphate, greenish-blue phosphorescence, without bands or lines. In the phosphoroscope, the colour is green, red and orange being cut off.

Chromium (5 per cent.) with lime, pale-reddish phosphorescence. The phosphoroscope cuts off red and orange.

Diamonds glowing pale-blue have the longest residual glow; next come those glowing yellow. Those phosphorescing red have no residual glow.

Glucina, rich blue; no residual glow.

Lanthanum sulphate, reddish. The spectrum shows a broad orange band with a sharp line-1/12 280-superposed. This is identical with the line Ge. Lime changes the colour to yellow, bringing out lines of yttrium and samarium, these being present as impurities. Gẞ is, however suppressed, together with a portion of the neighbouring spectrum.

« PreviousContinue »