s66 



NA TURE 



[October 8, 1896 



nnd J mm. per sec. (or of that order). There are extraordinary 

 differences between the rates of leakage in different gases ; 

 roughly they follow a density law. Thus mercury vapour (which 

 is one of the best insulators) is here found to become the best 

 conductor. Chlorine, bromine, and iodine come next. Sul- 

 jjhuretted hydrogen conducts better than oxygen. But the rates 

 depend on the slope of potential u.sed, and the order may even 

 be reversed (as in the case of air and hydrogen). Another 

 remarkable result of the ex|ieriments is to .show that the con- 

 ductivity under the action of X-rays increases when the length 

 of the column of gas between the electrodes is increased ; this is 

 intelligible on the electrolytic theory just referred to and is, 

 indeed, required by it. Prof. Ayrton pointed out that a similar 

 phenomenon is observed in arc lamps worked at a constant 

 potential ; when the length of the arc is increased, the current at 

 the same time increases. Prof. RUcker made a preliminary 

 communication on measurements of transparency of glass and 

 porcelain to Riintgen rays, made by himself and Mr. Watson. 

 A colour-patch photometric method was employed, in which the 

 light produced by the rays on a phosphorescent screen was com- 

 pared with light from an arc lamp which had passed through two 

 thicknesses of cobalt glass. Assuming the law of inverse square.s, 

 it is found for glass that the intensity of the transmitted light is 

 given by 1= Iq (A-I- B') where Iq is the intensity of the incident 

 light and / the thickness of the glass. Certain kinds of china are 

 almost as transparent as glass ; but others, in the manufacture 

 of which bone ash is used, are very opaque. This method of 

 examination may prove ot service to collectors of porcelain and 

 china. Lord Kelvin made a preliminary communication on 

 measurements (by himself. Dr. Bottomley, and Dr. Maclean) of 

 electric currents through air at different densities down to one 

 five-millionth of the density of ordinary air. .At a distance of 

 I '5 mm. between needle-points an E. M.F. below looo volts 

 produces no current. 2000 volts produces an appreciable 

 current which increases rapidly from 2000 to Sooo volts. A 

 curve having volts as abscissoe, and galvanometer leadings as 

 ordinates, is always convex to the axis of absciss-i;-. The above 

 measureinents were made at the ordinary pressure ; at a pressure 

 of one thousandth of an atmosphere (075 mm.) a few score of 

 volts would start a current. Dr. K. T. Trouton communicated 

 the results of experiments on the duration of X-radiation at each 

 spark, made by rotating a wheel between the discharge 

 lube and a sen.sitive plate. The duration varies between 

 l/300th and i/io,ooolh of a second, but the results are 

 dependent on the nature of the plate used. Prof. .S. P 

 Thompson read a paper on the relation between kathode rays, 

 X-rays, and Becquerel's rays. Interesting experiments were 

 described in which various .screens and obstacles were introduced 

 inside a Crookes tube. In one of these the discharge from a 

 concave kathode was allowed to fall on a flat anti-kathode 

 inclined at 45°, and then on to two aluminium wires as obstacles 

 in front of the wall of the bulb. At a low degree of exhaustion 

 kathode rays are produced which throw shadows of the wires on 

 the bulb, but no shadow on a fluorescent screen ouLside. The 

 position of the shadows on the bulb can be .shifted by a magnet. 

 At a high degree of exhaustion we get X-rays which throw 

 .shadows on a fluorescent screen outside. These are not shifted 

 directly by a magnet, excepting that the magnet shifts the hot 

 point of the kathode rays on the anti-kathode. At an inter- 

 mediate degree of exhaustion both shadows can be .seen 

 simultaneously. Tbe kathode shadows contract when the wires 

 are electrified positively, and expand when they are electrified 

 negatively ; the Xshadows are not affected by electrifying the 

 wires. 



On Saturday the Section divided into two departments. In 

 the department of Physics the Report of the Committee on the 

 Comparison and Reduction of Magnetic Observations was 

 presented. Prof. Riicker presented the Report of the Committee 

 on Magnetic Standards. A survey instrument previously compared 

 with Kew has been taken to three observatories and compared 

 with the instruments at these. Prof. RUcker visited Falmouth, 

 and Mr. W. Watson Valencia and Stonyhurst. The differences 

 from Kew are given below : 



Kalmuvuli. Stony)iursl. V.-iIencia. 



Declination -0-2 ... -)- 1 t ... +00 



H X io-»C.G.S. ... - iS ... -6 ... T 29 



Dip - i'-6 ... -)- 2'-2 ... - I'-S 



In the course of the adjourned discussion on Prof. S. P. 

 Thompson's paper (read on Friday), Prof. V. Bjerknes stated 

 that M. Birkelar.d (of Kristiania) had recently observed a dis- 



NO. 1406, VOL. 54] 



continuous line .spectrum of kathode rays produced by magnetic 

 deflection. The rays are allowed to pass through an aperture in 

 a metallic screen inside the tube, and their position, after 

 deflection, is observed by means of the fluorescence on the wall 

 of the tube. When the pressure is high only a single line is 

 observed, but when the pressure is reduced new lines make their 

 appearance. The spectrum is not continuous, as Lenard had 

 supposed, but is discontinuous like the line s])ectrum of a gas. 

 This sup|)orts the view that the rays are due to ethereal 

 vibrations. The observations are rather difficult on account of 

 flickering. Threeor four bright lincsare distinctly seen, but probably 

 there are thirty or forty present. I'rof. ,S. P. Thompson read a 

 further jtaj^er on '* hyper- phosphorescence " — the term hyper- 

 Ijhosphorcscent lieing applied to bodies which, after due 

 stimidus, exhiliit a persistent emission of invisible rays not 

 included in the hitherto recognised spectrum. In endeavouring 

 to shorten the time of exposure in photographing with X-rays, 

 the action of fluorescent substances, such as calcium sulphide, 

 zinc sulphide, and fluoride of uranium and ammonium, was 

 tested. The plates were found to become fogged by these 

 materials, although they had been kept in the dark long enough 

 for all visible jihosjihorescence to disappear. Even after being 

 kept in the dark for six weeks calcium sulphide actively emits 

 rays that affect a jihotographic plate. Experiments were made 

 to test whether sunlight, or the light from an arc lamp, contains 

 any radiation which will pass, like the .X-rays, through opaque 

 bodies. From an arc lamp, with an exposure of two hours, 

 photograjihic shadows of pieces of metal were obtained through 

 pine-wood several millimetres thick ; but aluminium was quite 

 opaque to everything radiated from the arc and to sunlight. 

 Fluorescent substances were next placed on top of an aluminium 

 plate below which was a photographic ]3late ; and the whole was 

 exposed to dull sunlight for several hours. Photographic actior> 

 was found to have taken place (through the aluminium) behind 

 the portions where uranium nitrate and uranium ammonium 

 fluoride had been placed. These effects are inconsistent with a 

 law enunciated by Stokes— but which he has since modified. 

 When they w ere communicated to Sir George Slokes, he drew the 

 speaker's attention to the remarkable results obtained by 

 Becquerel in the .same direction. The Becquerel rays differ from 

 the X-rays in that they can be refracted and polarised ; they are 

 probably transverse waves of an exceedingly high ultra-violet 

 order. 



In the department of Mathematics a Report was presented on 

 the G (;-, v) Integrals ; also the Report of (he Committee on 

 Bessel Functions and other Mathematical Tables. Papers were 

 read by the Rev. R. Harley, on results connected with the theory 

 of differential resolvents ; by Lieut. -Colonel .A. Cunningham, 

 on the connection of quadratic forms ; by .Mr. H. M. Taylor, 

 on great circle sailing ; by Mr. S. H. Burbury, on the stationary 

 motion of a system of equal elastic spheres : and by Mr. G. H. 

 Bryan, on some difficulties connected with the kinetic theory of 

 gases. 



On Monday the Section again met in two departments. In 

 the Physics Theatre, Lord Kelvin gave an account of experiments 

 made by himself. Dr. Maclean and Mr. Gait, on the communica- 

 tion of electricity from electrified steam to air : and also con- 

 tributed a paper on the molecular dynamics of hjdrogen gas, 

 oxygen gas, ozone, peroxide of hydrogen, vapour of water, liquid 

 water, ice, and quartz crystal. Taking hydrogen and oxygen 

 and their compounds first, it is assumed that there are two 

 kinds of atom, /;, », with the distinction that the force between 

 two Ks and the force between two o's and the force between 

 an // and an are generally different. The mutual force between 

 two h'i is always the .same at the same distance : .so is the force 

 between two o's and between an /; and an 0. The atoms are 

 considered as points acting on one another with forces in the 

 lines joining them ; no further assumption is made beyond the 

 conferring of inertia upon these Boscovich atoms. It is shown 

 that the "known chemical and physical properties of the sub- 

 stances named can be explained by making H consist of two- 

 Boscovich atoms (/;, h) and O of two others (,>, ,)). This makes 

 Hj consist of four //'s at the corners of an equilateral tetrahedron, 

 and O., a similar configuration of four o's. It naturally shows 

 ozone 'as six 0% at the corners of a regular octahedron ; and 

 peroxide of hydrogen as a tetrahedron of /;'s symmetrically 

 placed within a tetrahedron of c's. It makes H.jO (gaseous) 

 consist of two o's, with two h'% attached to one of them and two 

 other h\ to the other ; the h\ of each getting as near to 

 the other as the mutual repulsion of the //'s allow s. Models of 



