August i8, 1892] 



NATURE 



385 



other. To obtain the average difference of the quantity of elec- 

 tricity set in motion in one direction over that in the other, an 

 electrolytic cell and a vacuum tube were placed in the secondary 

 circuit of a small Ruhmkorff coil. The solution in the cell was 

 sulphate of copper of density 17 with \ per cent, of commercial 

 sulphuric acid added. The mean of seven experiments, lasting 

 from two to four hours, gave the average electrolytic current, 

 calculated from the gain of the cathode, as ^ of a milliampere. A 

 similar experiment, in which the vacuum tube was replaced by 

 a very large liquid resistance, led to no result. 



The Application of Interference Methods to Spectroscopic 

 Measurement, by Prof. A. Michelson. — Prof. Michelson's 

 "wave-compiler" consists essentially of a small plane sheet of 

 glass with parallel surfaces and two mirrors. The mirrors are 

 set at right angles to each other and the central plane of the 

 glass passes through their line of intersection, making an angle 

 of 45° with each of them. Rays of light from the source under 

 examination fall upon the glass surface at an angle of 45°, and 

 are partly reflected, partly transoiitted, so as to suffer normal 

 reflection at the mirrors, and finally proceed from the other 

 surface of the glass, at an angle of 45°, to the eye. Normal dis- 

 placement of one of the mirrors parallel to itself causes aidiffer- 

 ence of path which produces interference. If the source of light 

 emits radiation of two wave-lengths (as in the case of incan- 

 descent sodium), or of more, the brightness of the interference 

 bands — regarded from the centre outwards — exhibits periodic 

 variation, which can be accurately observed. The law of varia- 

 tion can be calculated when the distribution of light in the 

 source, as regards wave-length and intensity, is known. Con- 

 versely, the method can be used to determine the nature of this 

 distribution. Mr. Michelson has examined various sources of 

 light — oxygen, hydrogen, zinc, cadmium, mercury, &c. — and 

 has found that lines which in the most powerful spectroscope 

 appear single, are really double, triple, or even more complex. 

 In examining hydrogen at different pressures and temperatures 

 the results indicated that the widths of the component lines 

 decreased as the pressure decreased, but not without limit. In- 

 vestigations made on a large number of substances give strong 

 confirmation of the kinetic theory of gases. 



On a Periodic Effect which the Size of Bubbles has on their 

 Spccd of Ascent in Vertical Tubes containing Liquid, by Dr. F. 

 T. Trouton. — The chief peculiarity observed when a bubble of 



the instrument Prof. Ewing has observed true effects of magnetic 

 time-lag in the inward penetration of magnetization. 



On a Magnetic Balance, and its Practical Use, by Prof, du 

 Bois. — A test bar of standard size is placed within a magnetizing 

 coil. Over this is placed an iron yoke, balanced on a knife edge, 

 and having attached to it a graduated scale with sliding weights. 

 When a current passes through the coil the equilibrium is dis- 

 turbed, and it is restored by sliding the weight along the scale. 

 The position of the weight then gives the magnetization in abso- 

 lute measures in c.g.s. units. The position of the yoke is one of 

 unstable magnetic and mechanical equilibrium. 



On Earth Current Storms in 1892, by Mr. W. H. Preece. — 

 In communicating this paper, Mr. Preece spoke of the great im- 

 portance of observers at all parts of the globe contriving to collect 

 data regarding electric storms. 



On the Dielectric of Condensers, by Mr. VV. H. Preece. — The 

 author pointed out that in the condensers used by him there was 

 evidence of work done upon the insulating material, which 

 necessarily retarded the rate of propagation of signals. 



On Polarizing Gratings, by Prof, du Bois. — The author has 

 constructed minute gratings with silver wire scarcely visible to 

 the naked eye. Radiant heat and long light waves are polarized 

 by these gratings in the same way as electromagnetic radiations 

 are polarized by larger wire gratings. 



The Volume Effects of Magnetism, by Dr. C. G. Knott.— The 

 results for iron tubes have been already described in Nature. 

 In one case a steel tube, of given bore and thickness of wall, 

 gave increase of internal volume in all fields used. Usually the 

 volume diminishes in low fields and increases in high. The 

 effects were shown to the audience by projection upon a screen. 

 An Estimate of the Rate of Propagation of Magnetization in 

 Iron, by Prof. Fitzgerald. — Assuming that the iron is constituted 

 of a system of little magnets, and with possible assumptions as 

 to the size of these magnets and their strength, it is found that 

 their natural rate of vibration may be one hundred millions per 

 second. Unless the period of the vibration propagated through 

 the iron approximates to this the wave lengths would be very 

 small ; while quicker vibrations, with periods like those of light, 

 would not be propagated at all. 



Experimental Proof that the Coefficient of Absorption is not 

 affected by Density of Illumination, hyTiT. W. Peddie. — When 

 parallel rays of light pass through a uniform absorbing medium. 



air ascends in water is that the speed of ascent is a periodic | the intensity of the light diminishes according to a certain law 



function of the size of a bubble. The form of the curve ob- 

 tained by plotting the volumes of the bubbles as abscissae and 

 the corresponding speeds as ordinates showed that at first, as 

 would be expected, increase in size diminishes the speed ; but 

 afterwards the speed increases in value, then reaches a maximum 

 at about twice the minimum speed ; and so on two or three 

 times, depending on the diameter of the tube employed. The 

 oscillations in the curve die out in much the same fashion as 

 those of a pendulum in a viscous medium. The form of the 

 bubbles was almost spherical at the first minimum ; after this 

 the bubble is pointed at the top until the second minimum is 

 reached, when it is again rounded at the top, but has a dumb- 

 bell shape, and so on, presenting in this way similarities to the 

 breaking up of a liquid column through surface tension. Liquids 

 which do not mix with water were used instead of air ; and air 

 bubbles in other liquids were also used. 



On a Method of Determining Thermal Conductivities, by Mr. 



C. H, Lees. — The method has more direct application to the 



determination of the conductivity of a liquid than previous 



methods have. It consists in measuring the amount of heat 



conducted, under given conditions, through a film of liquid 



placed between two copper cylinders. It was found most con- 



nt to keep the upper cylinder at the temperature of the 



inding air, while the lower one was kept cool by water 



the mains. Precautions were taken against errors from 



tion, &c. 



Magnetic Curve Tracer, by Prof. Ewing. — The apparatus is 



;ned to plot mechanically the ordinary magnetization curve. 



curve is traced on a screen by a spot of light reflected from 



nor, which is subjected to two motions — one proportional 



,e magnetizing force, the other to the magnetization. These 



motions are communicated by means of the sagging of wires 



placed in air-gaps in magnetic circuits. In the one case, the wire 



carries a steady current in a varying field ; in the other the wire 



carries the varying current in a steady field. The curves may 



be traced on sensitized paper ; and the instrument should be of 



ttuch use to engineers for testing purposes. In working with 



The assumption which, on this point, is made the basis of the 

 theory of radiation is that the fractional dimunition of intensity, 

 at any stage, per unit of thickness traversed (called the coefficient 

 of absorption), is independent of the intensity of the light. Sir 

 G. Stokes has indicated a method of testing the point 

 by the reflection of light, at nearly perpendicular incidence, 

 from the surface of glass — ^part of the absorbing medium 

 being placed in the path of half of the light before 

 reflection, and a similar part being placed in the path of the 

 other half after reflection. Both portions, being then projected 

 on a screen, could be directly compared in respect of colour and 

 intensity. No test seems to have been made by this or any 

 other method. In Dr. Peddie's method light is passed through 

 two double image prisms and a plate of quartz. Four rays are 

 thus produced, coloured alike in pairs, the colour of one pair 

 being complementary to that of the other. The colour of one 

 pair can be made to match as nearly as possible the colour most 

 readily absorbed by the medium, similar portions of which are 

 placed at different distances from the points from which these 

 rays are made to diverge by means of a lens. The light being 

 projected on a screen, a direct comparison is obtained. The 

 media used were pieces of surface-coloured glass. In no case 

 was any difference observed, although the intensity varied from 

 I to 1000, and the eye could have observed the difference of 

 one per cent, in the brightness of the two discs thrown on the 

 screen, without the additional help of change of colour. 



On Dispersion in Double Refraction due to Electric Stress, 

 by Dr. John Kerr. — The fact of dispersion has been established, 

 and it is found that the optical effect depends upon the wave- 

 length, being the inverse ratio of the square root of the wave- 

 length. 



On a Delicate Calorimeter, by Messrs. J. A. Harker and P. J. 

 Hartog. This is essentially a Bunsen ice-calorimeter, with solid 

 acetic acid instead of ice, so being much more delicate, and 

 capable of being used at ordinary temperatures. 



On Graphic Solutions of Dynamical Problems, by Lord Kel- 

 vin. — The method of drawing meridional curves of capillary sur- 



NO. II 90. VOL. 46] 



