Stress-Strain Relations of Rubber. R. H. THURSTON.

(Science, vol. vii., 1898, pp. 522–523.)

The Author refers to some earlier work, and then says: "In all cases the substance behaved under load precisely as do other materials in the early part of its strain; then a reversed curve is described, and the test-piece stiffens greatly, and offers continually increasing resistance until at last rupture takes place, without yielding by inelastic deformation at any point of its course. Toward the end of its test the substance yields proportionately to the applied load. The fracture is sharp and without warning, and the break clean and smooth, and at right angles to the line of pull. No permanent reduction of section is observable after fracture. The reduced section immediately before breaking is but one-eighth the initial section of the unstrained rubber. Permanent set occurs to an exceedingly slight extent, and its value is dependent upon the maximum load, and independent of the elastic properties of the substance. The set of the material would not be noticed in ordinary use. Permanent loads produce permanent continuous extension, and in time fracture. This was found to be true for loads rising from 40 lbs. to 330 lbs. per square inch (2.8 to 23.18 kilograms per square centimetre), and stress-strain diagrams for two weeks showed steady elongation.

66

Plotting curves having for their co-ordinates loads per unit of area and areas of section of test-piece at point of maximum reduction, the stress-strain diagram thus produced becomes altered in form, and similar to those of other materials plotted in the usual manner. It has the same curvature at the initial stage, the same straight line to an (apparent) elastic limit, and finally a steady but slight rise with increasing loads, with a sudden break at the end. The highest load measured in these experiments was 810 lbs. per square inch (56.7 kilograms per square centimetre). The quality employed in all cases was that of the stationer's elastic bands."

The Author concludes with a reference to R. A. Fessenden's explanation of the observed phenomena.

A. Gs.

Corrected Value of Rowland's Measurement of the Mechanical Equivalent. W. S. DAY.

(Physical Review, vol. vi., 1898, pp. 193–222.)

The Author has compared three of Rowland's thermometers directly with three Tonnelot thermometers which had been completely studied at the Bureau International (two of which were broken in crossing the Atlantic). The Tonnelot thermo

meters were used exactly according to the directions given by Guillaume, in a specially constructed water-bath, which is described in detail. The Rowland thermometers could not conveniently be used vertically, as Rowland himself used them, but their pressure-coefficients were determined, and correction made for the difference of position. The treatment of them with regard to zero-point was the same as Rowland adopted. The thermometers, when corrected to the absolute scale according to Rowland's tables, showed differences from the Paris hydrogen scale amounting to 0° 032 as a maximum. Rowland's values for the equivalent are recalculated accordingly, and the results given for each degree from 6° to 36°. The following is an abstract :

Temperature.

6

10

15

20

25

30

35

The minimum occurs at 31°.5. J = 4.2 x 10 at 7°.5.

The temperature-coefficient agrees well with Griffiths' values from 15° to 25°. The absolute values are less than Griffiths' by 30, and less than Schuster's by ; the discrepancy is almost certainly not due to thermometric difficulties.

An abstract of the work occurs in the Philosophical Magazine, August 1897, but it contains an error.

R. A. L.

Mechanical Equivalent of Heat. J. B. BAILLE and C. FÉRY. (Comptes Rendus de l'Académie des Sciences, Paris, vol. cxxvi., 1898, pp. 1494–1496.)

A copper cylinder is placed in a rotating magnetic field, produced by surrounding the cylinder coaxially with a ring, to which, in a suitable manner, a two-phase current is supplied. The cylinder tends to rotate, but is prevented, the couple being measured by a balance. The frequency of the current is known, and the work spent on the copper can therefore be readily calculated. The rise in temperature of the cylinder is measured by a thermometer placed in a cavity, cooling corrections being applied. The values of the mechanical equivalent obtained vary from 422 to 426. The work is to be repeated with improved apparatus.

A. Gs.

Universal Shunt. J. RYMER-JONES.

(Electrical Review, vol. xlii., 1898, pp. 717-718.)

The Author, in view of the incorrect opinion that KelvinVarley slides are not applicable as a universal shunt, gives calculations showing that the scale-reading is proportional to the multiplying power of the shunt. He then describes a simplified form of Kelvin-Varley slides, in which the 100-vernier coils are fixed to a movable ebonite disk, to which also are fixed the springs making contact with the main resistances. For the rough adjustment the whole disk is moved round, and the same handle serves, on loosening a detent, for the fine adjustment with the vernier coils.

Conductivity of Gases. R. W. WOOD.

(Physical Review, vol. vi., 1898, pp. 165-166.)

G. H. BA.

A compact apparatus of the kind devised by Kuntz is described, for showing the conductivity of gases. A larger bulb containing the gas to be tested (or a vacuum) surrounds a smaller bulb containing ether and communicating with a nozzle. On immersion in boiling water, the ether is more or less rapidly evaporated according to the conductivity of the intervening gas, and the vapour will burn at the nozzle in a jet of corresponding height. The bad conductivity of a vacuum is well shown. Glass-working directions are given.

A. D.

Determination of Thermal and Electric Conductivities.

P. STRANEO.

(Roma, R. Accad. Lincei, Atti, vol. vii., 1898, pp. 197–202.)

A mathematical Paper showing how it is possible, by simultaneous observations, to ascertain both the thermic and the electric conductivities of a wire at high temperatures. The error due to slight mismeasurements of the points at which certain stationary temperatures are observed is very small.

A. D.

Cadmium Standard Cell. P. KOHNSTAMM and E., COHEN.

(Annalen der Physik und Chemie, vol. lxv., 1898, pp. 344–357.)

The Weston cadmium cell shows an irregularity in the temperature coefficient below 5°, which renders its use in the cold less satisfactory. The Authors investigate the cause of the irregularity, and discover it in the fact that the curve of solubility of the cadmium sulphate undergoes a sudden change at 15°, remaining stationary between that temperature and 20°. This is no doubt due to some change in the constitution of the salt, which under ordinary circumstances corresponds to the formula CASO, H2O. This change is analogous to the change of crystalline form undergone by sulphur at 95°. The Weston cell should only be used at temperatures above 15°. It then is much more stable than the Clark cell.

4

E. E. F.

Fifth Annual Report of the Committee on Atomic Weights. F. W. CLARKE.

(Chemical News, vol. lxxvii., 1898, pp. 239-242.)

Papers published during 1897 which bear on atomic weights are critically reviewed. The Papers considered refer to the atomic weights of carbon, nitrogen, chlorine, silver, aluminium, nickel, cobalt, tungsten, and cerium. A complete list of atomic weights is given which differs from that published in 1896 in the following instances-carbon 12.00, cerium=139 35, cobalt=58 99, when 0 = 16. Rummel has worked out relationships between the spectra of the alkali metals and their atomic weights, from which the latter may be calculated with fair accuracy.

T. E.

Electromotive Properties of Chromium. W. HITTORF.

(Annalen der Physik und Chemie, vol. lxv., 1898, pp. 320-343.)

The position of chromium in the electromotive series depends upon its chemical state. Fresh fractures of the metal are in an active state, corresponding to the oxide CrO3, and in that state the metal stands immediately above zinc, and precipitates other metals from their solutions. In the inactive state, corresponding

p. 75.

Proceedings of the Royal Society of Victoria (Australia), vol. x., part

to the oxide CrO, it is a noble metal, reduces no other metal from its salt solutions, and stands at the electro-negative end of the series next to platinum. Experiments with a large number of different electrodes and electrolytes show that chromium as an anode may, with the same electrolyte, assume each of its three grades, according to the solvent and the temperature. By modern processes the metal can be obtained in large quantities, completely

fused and free from carbon.

E. E. F.

Polarization and Electrolysis. DEL PROPOSTO.

(Bulletin de l'Association des Ingénieurs Électriques, Liége, 1897, pp. 36–57.)

The Author has investigated certain discrepancies which are found to occur between the observed electromotive force of polarization and the value determined by Kelvin's law, for various solutions. He describes some experiments in which the terminal potential difference and the polarization electromotive force were measured for gradually increasing currents, and obtained curves similar to curves of magnetization, i.e., two substantially straight lines connected by a curve. From this the Author draws the following deductions, viz. :-Assuming that the two ions of the electrolytic molecule are charged with electricity of opposite sign so as to form a kind of elementary electromagnet, then, under the action of the electric force, the small magnets will be turned with their similar poles to the same sides, and the electrolyte will form a kind of magnet which only presents free masses (? magnetism) at the surfaces in contact with the electrodes. The mean intensity of magnetization, which corresponds to polarization in condensers, will be equal to the mean density of electricity on the extreme faces of the electrolyte, and the intensity of the current corresponds to the intensity of the field in the phenomena of magnetization.

The Author also investigated the effect of introducing diaphragms of glass and metal into the liquid between the electrodes, the diaphragms extending to various depths in the liquid. He found that the increase of potential difference for a given current passing through the voltameter follows approximately the same law as the fall of pressure produced by throttling a water-conduit. The Author also discusses the "secondary electrolysis" which is set up when the diaphragm is made of conducting material.

C. K. F.