556 



PHYSICS, PROGRESS OF, IN 1902. 



and without the production of light, through the 

 gas, as through an electrolyte, so that the ob- 

 served strata and lines exactly represent the un- 

 derlying reality. 



lonization.J. A. McClelland (Cambridge Phil- 

 osophical Society Proceedings, February) de- 

 scribes experiments on conductivity produced in 

 gases by the action of incandescent metals in 

 which there are many indications of the presence 

 not only of ions produced from the molecules of 

 the gas* but also of ions actually given off from 

 the hot wire itself. Thus at a pressure of 1 milli- 

 meter or less, the current when the wire is nega- 

 tive may be 50 times what it is when the wire is 

 positive. Such a difference from atmospheric 

 pressure suggests that, at low pressure, there are 

 numerous negative ions from the wire itself. J. 

 S. Townsend (Nature, March 6) draws the fol- 

 lowing conclusions from various experiments: 

 The negative ions set free from a zinc plate when 

 ultra-violet light falls on it generate other ions 

 by collisions with molecules of air, carbonic acid, 

 or hydrogen. The negative gaseous ions thus gen- 

 erated have the same properties as those generated 

 from the zinc. Thus ions given off by the zinc 

 plate, and those of air, carbonic acid, and hydro- 

 gen are identical. The negative ions generated by 

 Rontgen rays in a gas are also the same, being 

 identical in mass, free path, and charge. Thus 

 it is possible to detach from the molecules of 

 gases negatively charged particles, which are 

 small compared with the molecules, and are the 

 same from whatever gas they are removed. F. 

 Campanile and G. di Ciommo (Elettricista, March, 

 1902) have investigated the discharging effect of 

 ionized air that has passed over a volatile liquid, 

 as compared with the same air after passing over 

 a non-volatile liquid. The results show that the 

 discharging effect of the ionized air is increased 

 by the vapor of the volatile liquid. Possibly the 

 mingled vapor has a greater conductivity than 

 the ionized air, or the mingled vapor may be ion- 

 ized. 



Electrification of Air by Glow Discharge. O. 

 Lehmann (Annalen der Physik, December, 1901) 

 concludes from experiments in this phenomenon 

 that convective transfer of electricity by glow 

 discharge at a point is not due to conductivity of 

 the air. He finds also that an electrically charged 

 body brought into a unipolar electrical wind does 

 not lose its charge if it is of higher and similar 

 potential, while in the opposite case it loses its 

 charge completely, and acquires the opposite one. 

 Positive and negative electrified air diffuse in the 

 same way, and the electrification can therefore 

 not be conditioned by the presence of free moving 

 electrons. The convective flow fills the space 

 around the point, but behind an insulating plate 

 perpendicular to the lines of flow, there is a space 

 free from electrified air. In a region in which 

 both positively and negatively electrified air is 

 present, the air is apparently conducting. Electric 

 winds proceeding from two oppositely charged 

 points combine only partially, a greater part es- 

 caping unchanged to the walls of the room. 



The Arc.C. Fery (Comptes Rendus, May 26) 

 has measured the temperature of the crater of the 

 voltaic arc, which represents, according to Violle, 

 the temperature of ebullition of carbon. By ex- 

 trapolation of results obtained from various physi- 

 cal properties, Violle obtained values ranjriii;: 

 from 3,500 to 4,100. The author, by application 

 of Stefan's law, reaches a result near the lower 

 of these limits, and has tested it with an improved 

 form of optical pyrometer, with which he gets the 

 values 3,867 and 3,897. This was higher than 

 the value deduced above, but a repetition of the 



observations with graphitic carbon gave concord- 

 ant values. At the temperature of ebullition,, 

 therefore, carbon does not behave as a perfectly 

 black body. 



Leakage. M. Mache (Vienna Academy, Decem- 

 ber, 1901) finds that leakage is directly propor- 

 tional to the potential of the charged body, other 

 things remaining the same. This is contrary to 

 the results of Elster and Geitel. In a closed ves- 

 sel, where the leak from the conductor in it is 

 tested under the same conditions, it is found 

 gradually to increase to about the fourteenth day, 

 after which it remains constant. The leakage is 

 proportional to the pressure of the surrounding 

 gas, and a rise in temperature from 16 to 60 is 

 without influence. 



Magnetism. Theory. W. Voigt (Gottingen 

 Scientific Association, 3, 1901) investigates analyt- 

 ically the possibility of obtaining from the theory 

 of electrons an explanation of paramagnetism and 

 diamagnetism. He assumes that the translation 

 velocities of the electrons are small in compari- 

 son with the velocity of radiation, and also, as a 

 preliminary hypothesis, that the electronic mo- 

 tions are undamped. It is then found that 

 changes in the velocities of the electrons in conse- 

 quence of the formation of magnetic fields will not 

 give rise to magnetic phenomena. The effects of 

 damping are then considered, and the author 

 shows that, onthis hypothesis, in a constant mag- 

 netic field, if the energy dissipated is supplied by 

 means of any completely irregular series of im- 

 pacts, so that the mean value of the energy tends 

 to a fixed limit, the body will exhibit paramag- 

 netic or diamagnetic properties, according to 

 whether the mean energy of the electrons after 

 the impacts is mainly potential or mainly 

 kinetic. 



Effect of Field on Electric Resistance. J. J. 

 Thomson (Philosophical Magazine,- March) con- 

 cludes that, on the theory that the electric cur- 

 rent in a metal is carried by charged particles 

 moving freely through the metal, the resistance 

 should be increased by a transverse magnetic 

 force. The opposite is the conclusion of Van Ever- 

 dingen, whose results are based on the assump- 

 tions that the corpuscles which carry the current 

 behave like a perfect gas ; that the collisions with 

 the surrounding molecules are similar to those 

 between hard elastic bodies; and that the corpuscle- 

 between two collisions is free from any force ex- 

 cept that due to the external field. Thomson: 

 states that the second assumption of Van Ever- 

 dingen is not likely to be true, but that a colli- 

 sion ought to be regarded as consisting of a deflec- 

 tion of the path of the corpuscle, due to the force 

 exerted on it by a molecule near to which it 

 passes ; on this supposition it is highly improbable 

 that the resistance should be diminished by the 

 field. As to the third assumption, as the corpus- 

 cles are highly charged, and within distances of 

 less than 10~ 7 centimeters of the molecules of the 

 metal, it is almost certain, Thomson says, that 

 the forces exerted on the corpuscle by surrounding 

 molecules are enormously greater than those due 

 to the external electric field, and that at the end 

 of its free path the corpuscle rushes into or p.i~t 

 the molecule with which it is colliding with a 

 velocity very large compared with that with 

 which it started. 



Stability. M. Ascoli (Nuovo Cimento, Jan- 

 uary) has investigated the conditions under which 

 magnetization is not affected by shock. A per- 

 manent magnet perfectly stable to shock may al- 

 ways be obtained by adjusting magnetizations 

 and demagnetizations. The percentage of perma- 

 nent magnetism which must be sacrificed in order 



