PHYSICS IN 1899. 



709 



duces mathematically the result that in a con- 

 stant magnetic field ions moving at high speeds 

 travel along the lines of magnetic force, while 

 those of lower speeds move in spirals. In the 

 discharge of electricity through gases the negative 

 ion has always been found to move faster than 

 the positive, and this accounts for many of the 

 differences between the phenomena at the posi- 

 tive and negative electrodes. In like manner 

 Prof. Thomson's results account for the difference 

 in behavior of the negative glow and of the posi- 

 tive column in a magnetic field, the glow follow- 

 ing the lines of magnetic force and the positive 

 column taking a spiral path. Bouty (Paris 

 Academy of Sciences, July 24) finds that when a 

 tube of rarefied gas is placed in an electrostatic 

 field there is a critical intensity of this field below 

 which the gas acts as a perfect dielectric and 

 above which it allows a discharge to pass. The 

 relation between this intensity and the pressure 

 is expressed by the formula 



/- A ( 



where f is the critical intensity, p the pressure, 

 and A, B, and C constants, of which B, and per- 

 haps C, are independent of the nature of the gas, 

 and A increases with its molecular weight. Mil- 

 lers (Wiedemann's Annalen, LXVIII, 2, p. 196) 

 finds, in experiments on the manner in which 

 the pressure of the gas between two plates affects 

 the current which passes when they are brought 

 to different potentials and Rontgen rays are al- 

 lowed to traverse the gas, that the moisture of the 

 gas does not affect either the potential or the cur- 

 rent intensity. Near the pressure of maximum 

 resistance the current intensity varies as the 

 square root of the pressure. Himstedt (ibid., 

 LXVIII, 2, p. 294) finds that a disk connected 

 with one pole of a Tesla transformer (the other 

 terminating in a point beyond sparking distance) 

 is charged positively, negatively, or not at all, 

 according to the distance between point and disk. 

 To explain this, it is assumed that the point 

 gives out more positive electricity than negative, 

 and that the negative electricity is capable of 

 being propagated to a greater distance, no matter 

 through what gas the discharge takes place, the 

 only difference being that the transition takes 

 place at different distances in different gases. 

 Ebert (ibid.. LXVII, 3, p. 608) has measured the 

 energy consumed in vacuum tubes, and finds that 

 it attains a minimum at a certain low pressure. 

 His experiments show also that the dielectric 

 strength of a gas increases with the valency of 

 its molecules and the smallness of its sphere of 

 influence, and support the view that the luminous 

 phenomena of the tubes are due to dissociation 

 caueed by impact of cathode rays. Villari (Nuovo 

 Cimento, IX, p. 77) finds that the cooling effect 

 on a platinum spiral in a gas produced by the 

 passage of electric sparks is increased with in- 

 tensity of the spark. Lessening the pressure 

 lessens the cooling effect. The cooling appears 

 to be due to the impact of cool gas particles pro- 

 jected from the spark terminals. Fomm (Wiede- 

 mann's Annalen. July) describes a new phenome- 

 non of the vacuum discharge seen when a vacu- 

 um tube is surrounded by wire rings at its ends. 

 At a certain exhaustion blue rings concentric with 

 those appear and inclose patches of positive light, 

 which gradually disappear until the blue light 

 fills the whole section. As the pressure continues 

 to lessen, the blue light leaves the glass walls and 

 expands lengthwise into a greenish-blue beam that 

 proceeds in the direction of the other ring, and 

 shows all the properties of cathode rays. Cantor 



(ibid., LXVII, 2, p. 481) finds that under certain 

 conditions vacuum discharges can be obtained 

 that are apparently continuous. A coherer shows 

 the presence of waves, but whether the wave sys- 

 tem is quite discontinuous or not the author has 

 not been able to determine. Moore (American 

 Journal of Science, VI, p. 21) treats vacuurn- 

 tube phenomena from the point of view of kinetic 

 theory, regarding their study as that of ele- 

 mentary action in electrical conduction. He 

 traces an analogy between the flow of heat from 

 a hot body and the motion of a discharge stream. 

 Sutherland (Philosophical Magazine, XLVI1, p 

 269) maintains that J. J. Thomson's hypothesis 

 that the cathode stream consists of atoms disin- 

 tegrated into their component particles of primi- 

 tive matter is not necessary. He believes that 

 the electron, or natural electric unit usually as- 

 sociated with the atom to form an ion, resembles 

 the atom somewhat and possesses inertia. The 

 cathode stream contains separate negative elec- 

 trons and some ions, the proportions varying ac- 

 cording to circumstances. Impact on a thin 

 aluminum window filters out the ions, and the 

 transmitted electrons form Lenard rays. Ront- 

 gen rays are not electro-magnetic pulses, but are 

 due to internal vibrations of the electrons, which 

 would be caused wherever cathode or Lenard rays 

 were absorbed, but which would always be 

 damped by an associated atom, and hence would 

 not accompany ordinary light. The author finds 

 that his hypothesis explains the general phe- 

 nomena of cathode, Lenard, and Rontgen rays. 

 If it is true, it should be possible to get also a 

 stream consisting of separate positive electrons. 

 These may possibly be identical with Becquerel 

 rays. In commenting on this paper, Prof. Thom- 

 son (ibid., p. 415) states that previously to his 

 adoption of his own hypothesis he had considered 

 and abandoned that of separate electrons. One 

 of the difficulties that attend it is the large pro- 

 portion of heat developed by impact of cathode 

 rays. Wehnelt (Wiedemann's Annalen, July) dis- 

 tinguishes two classes of cathode-ray shadows, 

 one produced by rays emerging from the cathode 

 normal to its surface and crossing each other at 

 a focus in case the cathode is curved, and the 

 other consisting always of upright shadows 

 thrown by rays parallel to the axis of the tube. 

 Ewers (ibid., LXIX, 1, p. 167) concludes that 

 Goldstein's " candle rays " consist not of projected 

 anode particles, but of positive ions of cathode 

 material. Berg (ibid., LXVIII, 4, p. 688), by per- 

 forating the anode instead of the cathode, ob- 

 tains phenomena converse to the canal rays, thus 

 showing that rays from one electrode are propa- 

 gated across gaps in the other. Swinton (Pro- 

 ceedings of the Royal Society, LXIV, p. 377 ) finds 

 that the rays that in a focus tube proceed from 

 the front surface of the anticathode are reflected 

 cathode rays. This reflection is largely diffuse, 

 but is partly true specular reflection, the maxi- 

 mum of the reflected field being always that the 

 point where the angles of incidence and reflection 

 are equal. Guglielmo (Nuovo Cimento, IX, p. 

 131) finds that a filament of glass rendered fluores- 

 cent by cathode rays appears blurred when looked 

 at through a thick stream of the rays at right 

 angles to their course. This he believes to indi- 

 cate perturbations in the ether along the course 

 of cathode rays. 



Rontijen Rays (see also paragraph just pre- 

 ceding). Malagoli and Bonacini (Nuovo Cimento, 

 IX, p. 279) conclude that substances may be di- 

 vided into three groups: (1) Simple diff users of 

 incident Rontgen rays; (2) transformers of the 

 rays; and (3) bodies acting partly in both these 



