July i, 1909] 



NA TURE 



C after a time of charging t is given by the formula 

 C = C„(i + B log (! + /)()), where C„ denotes the capacity 

 when / = o and B and /> are constants. In the case of 

 ebonite at 30° C. this formula represents the results 

 obtained to within i part in 2000. The values of the 

 constants have been found for several substances at 

 different temperatures. The capacity is shown to be in- 

 dependent of the potential difference within the limits of 

 error. It is shown that after the temperature of an 

 ebonite condenser has been changed, then a very slow 

 change in the capacity goes on which continues for more 

 than 100 hours at constant temperature. — The effect of 

 pressure on the band spectra of the fluorides of the metals 

 of the alkaline earths : R. Rossi. It was shown by A. 

 Dufour that the band spectra of the fluorides of the 

 alkaline earths show a marked Zeeman effect, and it was 

 thought interesting to see whether these particular bands 

 would also be displaced by pressure, for it is known that 

 the cyanogen bands, which, like most bands, do not show 

 a Zeeman effect, are not displaced by pressure. The 

 large 2iJ-feet concave grating spectrograph of the 

 physical laboratory of the Manchester University was used, 

 and (he bands of the fluorides of calcium, barium, and 

 strontium were found to be shifted by pressure. The 

 order of magnitude of the displacement is about the same 

 as for line spectra. — The components into which the bands 

 are resolved are widened by pressure, and the linear 

 relation between pressure and displacement found by 

 former observers on line spectra seems to hold also for 

 these bands. There does not seem to be any evident 

 relation between the magnitudes of the Zeeman and 

 pressure-shift effect in the case of these bands. — The 

 ionisation produced by an o particle : Dr. H. Geigrer. 

 The aim of the experiment was an accurate determina- 

 tion of the number of ions produced by an a particle when 

 completely absorbed in air. The most direct way to find 

 the number of ions would be to measure the whole ionisa- 

 tion produced by the a particles from a known quantity of 

 radium C. Since it is, however, practically impossible to 

 obtain the saturation current due to the a particles at 

 atmospheric pressure, it was necessary to adopt an indirect 

 method. This method was brieflv as follows : — The 

 ionisation due to the whole number of a particles e.xpelled 

 from a known ciuantity of radium C was measured at a 

 low pressure, allowing only a small definite portion of 

 the range of each a particle to be effective. The ratio of 

 the ionisation produced within this small portion of the 

 range to the ionisation produced along the whole path 

 was then found from an accurate determination of the 

 ionisation curve. It was found that the number of ions 

 produced in air by an a particle from radium C along its 

 whole path is 2-37x10". Since the a particles from 

 different radio-active products differ only in their initial 

 velocity, it was possible bv the aid of the ionisation curve 

 of radium C to calculate the number of ions produced 

 by the other products. — .\ diffuse reflection of a particles : 

 Dr. H. Geigrer and E. Marsden. It was observed that 

 a diffuse reflection takes place when a particles are in- 

 cident on a plate. The reflected particles were counted by 

 the scintillations produced on a zinc sulphide screen. The 

 effect was found to vary with different metals as re- 

 flectors, the amount of reflection being approximately pro- 

 portional to the atomic weight of the reflecting substance. 

 L"sing different numbers of thin gold foils as reflectors, it 

 was found that the reflection was a volume effect, and 

 thus similar to the reflection of ;3 particles. Taking a 

 measured quantity of radium C as source, and using a 

 plate of platinum as reflector, it was found that, of the 

 incident o particles, about i in 8000 suffers reflection. — 

 The decay of surface waves produced by a superposed 

 layer of viscous fluid : W. J. Harrison. An estimate is 

 obtained of the effect of a thin layer of viscous liquid 

 on the decay of waves at the surface of a slightly viscous 

 liquid. The period equation for the motion is of the 

 fourth degree, and has two real and two complex roofs 

 in the case of waves of less than a certain length, and 

 four complex roots in the case of waves of greater length. 

 The real roots correspond to dead-bent modes, the complex 

 roots to propagated modes. No general expression of any 

 use can be obtained for the damping, but the equation 



NO. 2070, VOL. S^'i 



can be solved numerically in any particular case. In the 

 paper the velocity of propagation and the modulus of 

 decay are given for waves of length 2, s. ^^< 20 cm. at 

 the surface of mercury on which is superposed a layer 

 of glycerine i mm. in depth. An estimate is also obtained 

 for the damping when the wave-length is small compared 

 with the depth of the layer. Two other problems in the 

 decay of surface waves are discussed. — The passage of 

 electricity through gaseous mi.xtures : E. M. Wellisch. 

 (i) An experimental method (based on Langevin's method) 

 has been devised in order to ascertain whether there are 

 two distinct mobilities for the positive or for the negative 

 ions produced by Rontgen rays in a mixture of two gases, 

 or of a vapour and a gas. {2) No evidence was found 

 of the existence of the two distinct mobilities ; accordingly 

 it is necessary to conclude that the motion of the ion 

 through the medium must involve a mechanism of a 

 character such as to produce a statistical average. 

 (3) Experiments were conducted with regard to the effect 

 produced on the ionic mobilities in air by adding small 

 quantities of vapours. The mobilities showed a marked 

 decrease on the addition of alcohol and acetone, but were 

 not sensiblv affected by the addition of the heavier vapours 

 of methyl 'iodide and ethyl bromide. (4) Experiments were 

 performed with regard to the ionic mobilities in mixtures 

 of a gas and a vapour, the ions being formed from the 

 latter constituent only. As a result of the experiments, it 

 was shown that there must be, at all events initially, 

 a transference cf the charge (both positive and negative) 

 from the vapour to the gas molecule. (5) Experiments 

 were performed with regard to the st.ability of the vapour 

 ions in the presence of hydrogen ; it was shown that the 

 vapour molecules can accompany the charge to an appreci- 

 able extent, even in the presence of a considerable quantity 

 of hydrogen. (6) The mechanism by which the trans- 

 ference of charge from one molecule to another is effected 

 has been discussed ; there is reason to believe that the 

 transference takes place by the medium of a detachable 

 unit of positive electricity. (7) From the experimental 

 results a theory of the mechanism underlying the passage 

 of electricity through gases at ordinary temperatures and 

 pressures has been deduced. — k study of the use of photo- 

 graphic plates for the recording of position : Dr. C. E. K. 

 Mees. — The coeflicients of capacity and the_ mutual 

 attractions or repulsions of two electrified spherical con- 

 ductors when close together : Dr. A. Russell. The com- 

 putation of the electrostatic energy of two spherical con- 

 ductors when close together is an important problem in 

 spark systems of wireless telegraphy. In this case the 

 formula; previouslv given for the capacity coeflicients are 

 verv laborious to 'evaluate. By extending a mathematical 

 theorem due to Schlomilch, an approximate formula is 

 obtained for the sum of a certain infinite series. By 

 using this theorem, it is shown that when the spheres 

 are close together the ordinary series formula? for the 

 capacitv coeflicients can be written in forms which can 

 be rcadilv computed to any required degree of accuracy. 

 The author has re-computed and extended in this way 

 Kelvin's table for the capacity coeflicients of two equal 

 spheres when the least distance between them does not 

 exceed half the radius of either. When the spheres are 

 at microscopic distances apart, the formula become very 

 simple. Kelvin's table also for the rates at which the 

 capacity coefTicicnts of two equal spheres alter with the 

 distance between them, when this distance does not exceed 

 half the radius of either, has been re-computed and ex- 

 tended. When the spheres are very close together the 

 laws of attraction and repulsion are simple. Let the 

 radius of each sphere be a, let .v denote the least distance 

 between them, and suppose that the ratio V,,'V. of the 

 potentials of the two spheres is not nearly equal to unity, 

 and that -v/o is verv small compared with unity. In this 

 case the mutual force between the spheres is attractive, 

 and is given by 



"' ' ~ V- approximately. 



ox 



If the potentials of the spheres be equal, the repulsive 

 force between thon is, to a first approximation, given bv 

 Kelvin's formula for the repulsive force between two equal 



