313

General and Physical Chemistry.

Distinction between Spectral Lines of Solar and Terrestrial Origin. By M. A. CORNU (Phil. Mag., 22, 458-463).—The dark lines of the solar spectrum are of two kinds. Those of one class (C, F, D, E, G, H, and the magnesium band b in the green) are of solar origin, and always present the same aspect: others (A, B, and band a in the red) are telluric, produced by selective absorption in the earth's atmosphere, and become broader and darker as the sun sinks towards the horizon. It is by this latter characteristic that they have been chiefly recognised hitherto.

The author's method of distinction is based on Fizeau's principle of the displacement of lines in the spectrum, when the source of light is in a state of absolute or relative motion. This displacement may be connected with the velocity by very simple expressions, and has been used to calculate the velocity of the source. In practice, the method consists in allowing the light from opposite edges of the solar disc (where of course the motions are opposed) to fall on the slit of the spectroscope; the telluric lines remain steady, while the solar lines appear to move. The motion is easily detected by noting the position of any line with respect to one of the particles of dust always found on the cross-wires. It becomes still more sensible when, by an arrangement which is figured and described, the collecting lens is caused to oscillate two or three times a second, so as to throw opposite edges of the sun's image alternately upon the slit. The solar lines are then distinguished at a glance.

The method has been applied to the anatomy of band a, and to ascertaining the telluric origin of some lines beyond &, and the solar origin of Kirchhoff's line 1474. Since this line oscillates, the vapour producing it must be carried round by the sun's rotation.

CH. B.

Spectrum of Germanium. By G. KOBB (Ann. Phys. Chem. [2], 29, 670-671).-The author has examined the spectrum of germanium by viewing in a six-prism spectroscope the spark of an inductioncoil taken between terminals of platinum and germanium. The

position of each line was determined by micrometric measurements of its distance from two neighbouring solar lines. The results are contained in the table (p. 313), in which wave-lengths are given in

1

1010

metre.

A. H. F.

Fluorescence of the Pigments of Fungi. By A. WEISS (Chem. Centr., 1886, 670-671).—The fluorescence of alcoholic extracts of fungi was examined by means of a cone of light. All gave a greater or less fluorescence, which was green with yellow or brown-coloured fungi, and blue with red or violet-coloured fungi. The ochre-yellow colouring matters of some Agaricinæ give, however, a sky-blue, and the red colouring matter of the heads of Ammanita muscaria, a green fluorescence. The spectrum of the blue fluorescent colouring matters of Russula shows a wide, very characteristic, black absorption-band in the green and yellow, a feeble one between the lines E and F, and a total absorption of the violet to the line G. The green-yellow band agrees in position with the band which is seen in the spectrum of a living red peony leaf, and also with that given by the blue colouring matter of many Campanulæ, after treatment with sulphuric acid. The more intense the colour of the extract, the more the absorption extends towards the red, so that with a thick layer of liquid the whole of the green and yellow is absorbed. The absorption in the violet is similar to that given by the red, blue, and violet leaf colouring matters of the Phanerogama. The green fluorescent colouring matters of fungi show a feeble absorption-band between E and F, and a wide absorption of the violet end of the spectrum, sometimes even the entire end of the spectrum as far as b is absorbed.

G. H. M.

New Secondary Element. By M. KALISCHER (Phil. Mag., 21, 164). The element consists of iron (or carbon) and amalgamated lead in contact with mercury, in a solution of lead nitrate. When charged, the iron (as anode) becomes passive and coated with lead peroxide, which protects it from the liquid. During discharge the peroxide is reduced. Electromotive force on open circuit = 2 to 2.5 volts; on closed circuit = 1.8 volts, falling slightly after a time. The lead plate must be occasionally renewed. Cн. B.

Electromotive Force of a Constant Voltaic Cell with Moving Plates. By A. P. LAURIE (Phil. Mag., 21, 409-415).— Before accepting measurements of electromotive force it must be shown, by analysis of the products found in the cell, to what reactions it is really due. This can only be done when the current is passed for a long time, since the electromotive force observed by an electrometer, or by connecting for a short time through a highresistance galvanometer, may be really due to impurities in the metal, or to a film of gas or oxide on its surface. Hence may arise the want of agreement in some cases between the electromotive forces observed by Wright and those calculated by him from the thermal data.

It is well known that the fall in electromotive force of a cell when the circuit is closed, due to alterations in the layers of liquid in con

tact with the plates, may be prevented by keeping the liquid in motion. The author has measured the electromotive force of cadmium and platinum immersed in iodine solution, by means of a galvanometer, and finds that it remains constant for a long time when the cadmium plate is kept moving by clockwork. Its initial value was 1.084 volts (by the electrometer 1076 volts). After two hours, it fell to 1.067 volts (by the electrometer 1.072 volts).

CH. B. Electromotive Force of Voltaic Cells having an Aluminium Plate as Electrode. By A. P. LAURIE (Phil. Mag., 22, 213–216). -Wright (Abstr., 1885, 721) has found that the electromotive force (0-538 volt) of zinc-aluminium cells (zinc in zinc sulphate, aluminium in potash alum) is opposite in direction to that calculated from the thermal data (0·938 volt); so also for other aluminium cells. The author attributes these contradictory results to the well-known property of aluminium in contact with air or water of becoming coated with oxide; and states that the abnormal electromotive force (measured by an electrometer) is reduced to 0.14 volt on cleaning the aluminium with sand-paper. When the aluminium plate is amalgamated, the electromotive force becomes normal, and equal to 046 volt; and the plate is speedily covered with a growth of oxide. Two aluminium wires, one cleaned, the other amalgamated, placed in a solution of aluminium sulphate, give an electromotive force = 1.08 volts.

Wright in reality measured the electromotive force between aluminium oxide on an aluminium plate and zinc; and the value obtained was probably due to the heat of formation of zinc sulphate--that of aluminium sulphate + that of aluminium oxide that of water. CH. B.

Electrical Resistance of Soft Carbon under Pressure. By T. C. MENDENHALL (Phil. Mag., 22, 358–363).—The author describes experiments on this much disputed point, which prove, in his opinion, that a decided diminution occurs in the resistance of carbon when it is submitted to pressure, independently of any change in surfacecontact between the carbon and the electrodes through which the current is introduced. For hard carbon, the change is slight, but greater that can be accounted for by any heating effect.

A disc of soft carbon, such as that used in Edison's tasimeter, showed a very great diminution in resistance when compressed between two columns of nercury. A pressure of 5 mm. of water caused a decided deflection in a sensitive galvanometer in circuit; and a pressure of 50 mm. of mercury reduced the resistance to one-half. If the initial pressure was considerable, the disc only slowly recovered its normal resistance after its removal.

Since hard carbon is much more porous than soft, the change should have been less with the latter than with the former, were the action entirely at the surface.

CH. В.

Electrolysis of Silver and of Copper, and the Application of Electrolysis to the Standardising of Electric Current and

Potential Meters. By T. GRAY (Phil. Mag., 22, 389-414).-This paper is chiefly of physical importance. The following points are, however, of general interest.

When the highest accuracy is required in electrochemical measurements, silver solutions are to be preferred to copper solutions; for although silver obtained by electrolysis is apt to be less coherent and less easy to manipulate than copper, it is not liable to oxidation and corrosion by the liquid. In ordinary measurements, however, copper is preferable.

In all cases thorough cleaning of the plates is of supreme importance; the best methods of effecting this are described at considerable length.

The size of the plates may vary within moderate limits for silver, and within wide limits for copper, without affecting the quality of the deposit, and rendering the processes of washing and weighing more difficult. For silver electrolysis, the author recommends a 5 per cent., or at most 10 per cent., solution of silver nitrate, and a cathode plate of not less than 200, nor more than 600 sq. cm. of surface per ampère of current. When the cathode is too small, the deposited metal is not adherent, and tends to grow out in crystals, especially from any sharp edge or corner. An anode plate slightly larger than the cathode is recommended; it should have a surface of at least 400 sq. cm. per ampère. Very small anode plates become black and spongy, and gas is evolved from the surface; but when the plates are large they remain bright, and may be weighed as a check on the weight of the cathode. When a silver anode has not been properly cleaned, the outer layer of metal is not dissolved, but remains as a loose skin. Between successive experiments, silver plates should be heated to redness in a spirit flame.

For the electrolysis of copper sulphate, the cathode plate should have a surface of at least 20 sq. cm. per ampère for short experiments, or 50 for experiments lasting some hours. The anode should expose at least 40 sq. cm. of surface per ampère; but the minimum size depends greatly on the degree of saturation of the copper solution. When this plate is too small, the passage of the current may be completely stopped after a time, owing to the metal becoming coated either with finely crystalline copper sulphate or with oxide. The loss from the anode never exactly equals the gain of the cathode; but when the current density does not exceed th of an ampère per sq. cm., the anode may be weighed as a control. It never becomes black and inelastic like a silver anode; in other respects the two behave in a similar way.

The density of the copper sulphate solution may vary between 1.05 and 1.18. With weak solutions, the deposited copper is not very coherent. A source of error in using copper lies in its corrosion by the liquid. This error is always trifling, and is a minimum for densities between 11 and 1:15. Copper plates are more readily corroded by neutral sulphate of copper solution than by a solution containing even 5 per cent. of sulphuric acid; and the action appears to be greater in a cell when the current is flowing that when it is not. A plate of copper immersed in pure sulphate solution at first loses weight by corrosion;

after a time it increases in weight owing to oxidation, and the increase finally becomes rapid owing to the formation of hydrated oxide.

The author gives 0.0003287 gram as the true amount of copper deposited by one coulomb of electricity. But for simple measurements of current the value may be taken as 0.000329 when the cathode has a surface of 50 sq. cm. For other sizes, this number may be corrected by means of a curve which is given. Direct determinations of the constant for silver were not very successful; but they seem to confirm the results obtained by Kohlrausch and by Lord Rayleigh. Сн. В.

Electrolytic Polarisation produced by Small Electromotive Forces. By C. FROMME (Ann. Phys. Chem. [2], 29, 497-544).— The polarisation of platinum plates in dilute sulphuric acid was measured during the flow of the polarising current, which was always so small as to produce no visible evolution of gases. Two forms of electrolytic cell were employed, the first open, containing two platinum electrodes in cells, joined by a tube containing a third, and a second form in which there were four equidistant electrodes contained in a glass tube. The acid employed varied in strength from 1 per cent. to 3 per cent., the amount of dilution not being found to produce any observable result. With the latter form of cell, the two middle electrodes were joined to the battery through a variable resistance, the two outer plates being for the purpose of measuring the potentials of the others by connection with a quadrant electrometer. In the first set of experiments, it was found that the maximum polarisation had occurred by the time the electrometer could be read, and the difference of potential was nearly equal to that obtained by joining the battery directly to the electrometer. Experiments with the open voltameter led to the conclusion that the difference of potentials between the plates remained nearly constant from the instant of making the current, but that the polarisation of the oxygen plate continuously increased, whilst that of the hydrogen plate decreased. In a voltameter free from air, the changes were similar, but in an opposite direction. The influence of a previous polarisation of the electrodes was examined, the battery being joined first in the same direction, and secondly in the reverse direction to that of the former current, and also the effect of polarisation of only one of the electrodes, this being effected by introducing temporarily a third electrode, and using it as an anode or cathode as required. By raising an electrode gradually out of the liquid, the effect of change of area was examined, and it was found that if either anode or cathode were raised so as to lessen its area of immersion, its polarisation was increased at the expense of the other.

Many other similar experiments were made, but without leading to any general conclusions, the author intending to extend the investigation, using gold and silver electrodes. A. H. F.

Expansion of Mercury between 0° and -39°. By W. E. AYRTON and J. PERRY (Phil. Mag., 22, 325-327).-By comparing the indications of a mercurial thermometer with those of a specially