710 



PHYSICS IN 1899. 



ways. These results differ from those of Sagnac, 

 who would place all bodies in the second group. 

 He asserts that there is no diffusion of .Rontgen 

 rays; that the rays transformed by different 

 bodies differ in penetrative power and absorbabil- 

 ity by air; and that after transmission the rays 

 are always accompanied by some transformed 

 rays. The same investigators (Atti dei Lmcei, 

 VIII p. 296) believe that there is no true de- 

 flection of Rontgen rays, the apparent deflection 

 being due to diffusion or to secondary rays by 

 transformation. Various observers have at- 

 tempted, in the absence of regular reflection and 

 polarization, to obtain the wave length of the 

 rays by diffraction. Maier (Wiedemann s An- 

 nalen LXVIII, 4, p. 903) obtains one thirtieth of 

 the wave length of ordinary light, which agrees 

 with Fomm's result. It is not certain, however, 

 that the "fringes" used by these investigators 

 are true diffraction phenomena. Haga and Wind 

 (ibid., p. 884) regard them as an optical illusion. 

 Zeleny (Proceedings of the Cambridge [England] 

 Philosophical Society, X, p. 14) finds that the 

 passage of electricity through a gas under the 

 influence of Rontgen rays is accompanied by con- 

 siderable motion in the gas itself, which for a 

 given intensity of the rays is dependent on the 

 potential gradient. The gas is dragged by the 

 moving ions in both directions equally if there 

 be no free charge, otherwise more in one direc- 

 tion. Experiments by Winkelmann on electric 

 currents induced by the rays are in accordance 

 (Wiedemann's Annalen, LXVI, 1, p. 1) with the 

 theory that the air between the plates is ionized, 

 and that almost as soon as the rays start a con- 

 stant electro-motive force exists between them 

 analogous to contact difference of potential. 



Electrification. Pellut (Comptes Rendus, 

 CXXVIII, p. 169) has shown that electrified 

 water lose part of its charge by evaporation. 

 The author applies his results to explain the daily 

 variation of atmospheric electricity after sunrise 

 and sunset. Kelvin, Maclean, and Gaet (Philo- 

 sophical Transactions, CXCI, p. 187) find that 

 when air in a glass cylinder is electrified by needle 

 points at 5,000 volts shorter times give larger 

 values. Five minutes suffice to annul the electri- 

 fication entirely. The authors suggest that con- 

 tinued electric disruptive action renders the gas 

 conductive. Blondlot (Comptes Rendus, CXXVIII, 

 p. 901) finds that an electro-motive force may be 

 developed by mixing liquids in a magnetic field. 

 The difference of potential disappears when the 

 stirring stops, and also when the magnetic field 

 is removed. If a person be supposed to lie on his 

 right side in the vessel facing the north-seeking 

 pole of the magnet, the positive electrode is at 

 his head. The expression for the electro-motive 

 force due to induction between the terminals is 

 an integral, containing, among other things, the 

 variable conductivity of the solution. 



Peltier Effect. Pochettino (Atti dei Lincei) has 

 investigated variations of the Peltier effect in a 

 magnetic field. Its coefficient is found to vary 

 with the magnetization, increasing up to a field 

 of 98 units, and then decreasing, passing its nor- 

 mal value at 345 units. This variation is inde- 

 pendent of the direction of magnetization. 



Resistance. Coloman de Szily (Journal de 

 Physique, June), from his experiments on the 

 effect of torsion on the resistance of wires, using 

 the alloy " constantan," whose resistance is but 

 slightly affected by change of temperature, draws 

 the following conclusions: Torsion increases the 

 electric resistance of a wire; up to the limit of 

 elasticity the increase is roughly proportional to 

 the angle of torsion, but beyond it the resistance 



increases more rapidly. The resistance of a 

 twisted wire decreases slowly with the time. 



Elvvtrulyxlx. Trouton (Electrician, Aug. 18) 

 finds that during electrolysis liquid passes through 

 a hole in a pierced partition between the elec- 

 trodes. The flow is independent of the current, 

 but is always toward the side of the hole where 

 bubbles are formed with greatest difficulty. Behn- 

 Eschenburg (Zeitschrift ftir Elektrochemie, V, p. 

 402) asserts that electrolytic conduction is sub- 

 ject to the same laws as metallic conduct i in. 

 basing his conclusion on experiments with a con- 

 tinuous nonconducting trough placed around the 

 iron core of a transformer and filled with a solu- 

 tion of an electrolyte. The trough constitutes an 

 elementary coil, in which the varying magnetic 

 field of the transformer induces currents. 

 Polarization. Scott (Wiedemann's Annalen, p. 

 388), from researches on polarization of thin 

 metal membranes, asserts that all such phenom- 

 ena in the case of gold are due to pores in the 

 metal. In platinum, however, they are caused by 

 occluded gases. The polarization capacity of sil- 

 ver and lead is affected by the metallic ions 

 in the electrolyte. Chassy (Comptes Rendus, 

 CXXVII, p. 1203) finds that the initial capacity 

 of polarization in an electrolytic cell is independ- 

 ent of the pressure. Cailletet had previously 

 shown that the capacity after passage of the cur- 

 rent is enormously increased by pressure. An- 

 thony (Proceedings of the American Association, 

 XLv II, p. 138) concludes that the high initial 

 electro-motive force of a cell is due to dissolved 

 oxygen, and that polarization is due to exhaus- 

 tion of oxygen in the electrolyte. 

 ' Thermo - elect ricit y Liebenow ( Wiedemann's 

 Annalen, LXVIII, 2, p. 316) shows that if \\< 

 assume that heat and electricity are so connected 

 that thermo-electric energy is produced win-n- 

 ever heat passes from a point at a higher to a 

 point at a lower temperature, every eli-Hrir ur 

 rent must be accompanied by a thermal current. 

 and therefore an electro-motive force in a metal 

 may force a stream of heat from a lower to a 

 higher temperature; in other words, thermo-elec- 

 tric phenomena are reversible. The author's fig- 

 ures with regard to the ordinary thermo-electric 

 current and the Thomson and Peltier effects 

 agree with experimental data. 



Dielectric* Edser (Proceedings of the Royal 

 Society, LXIII, p. 374) believes that if a definite 

 potential difference be maintained between any 

 two parallel planes in a dielectric medium two 

 kinds of molecular strain are produced molecu- 

 lar rotation and a separation of constituent 

 atoms. Double refraction is explained on the 

 assumption that the molecules have their axes 

 parallel, so that electrical disturbances perpen- 

 dicular to these will produce rotation and those 

 parallel will produce interatomic separation. 

 Thomas Gray (American Association), in obser- 

 vations on the dielectric strength of oils, points 

 out the peculiar fact that the strength per centi- 

 metre decreases as the thickness ot the hiyer 

 increases. His observations were made between 

 spherical terminals. Pellat (Comptes Rendus. 

 CXXVIII, p. 1312) explains residual dun-, by 

 supposing that the polarization of a dielectric 

 in an electric field does not take place instantane- 

 ously, and does not vanish instantly when the 

 field is abolished. He believes that he has experi- 

 mentally confirmed this hypothesis. 



Hertzian Waves. Le Bon (Revue ScientifKpie, 

 XI, p. 513) notes that metals appear not to l>e 

 as opaque to Hertzian waves as was at first 

 thought. If the inclosure be perfect, however. 

 the waves do not get through, for the apparent 



