OCTOBBE 17, 1902.] 



SCIENCE. 



629 



classified as follows: 'The Thermo-Electric 

 Powers of Pure Metals'; 'The Magnetic 

 Properties of Iron and Steel'; 'Dielectric 

 Constants'; 'The Magnetic and Electric 

 Constants of Liquid Oxygen ' ; ' Magnetic 

 Susceptibility. ' 



The investigations have shown that elec- 

 tric conductivity in pure metals varies al- 

 most inversely as the absolute temperature 

 down to minus 200 degrees, but that this 

 law is greatly affected by the presence of 

 the most minute amount of impurity. 

 Hence the results amount to a proof that 

 electric resistance in pure metals is closely 

 dependent upon the molecular or atomic 

 motion which gives rise to temperature, 

 and that the process by which the energy 

 constituting what is called an electric cur- 

 rent is dissipated essentially depends upon 

 non-homogeneity of structure and upon 

 the absolute temperature of the material. 

 It might be inferred that at the zero of 

 absolute temperature resistance would 

 vanish altogether, and all pure metals be- 

 come perfect conductors of electricity. 

 This conclusion, however, has been ren- 

 dered very doubtful by subsequent obser- 

 vations made at still lower temperatures, 

 which appear to point to an ultimate finite 

 resistance. Thus the temperature at which 

 copper was assumed to have no resistance 

 was minus 223 degrees, but that metal has 

 been cooled to minus 253 degrees without 

 getting rid of all resistance. The reduc- 

 tion in resistance of some of the metals at 

 the boiling-point of hydrogen is very re- 

 markable. Thus copper has only one per 

 cent., gold and platinum three per cent., 

 and silver four per cent, of the resistance 

 they possessed at zero C, but iron still re- 

 tains twelve per cent, of its initial resist- 

 ance. In the case of alloys and impure 

 metals, cold brings about a much smaller 

 decrease in resistivity, and in the case of 

 carbon and insulators like guttapercha, 



glass, ebonite, etc., their resistivity steadily 

 increases. The enormous increase in re- 

 sistance of bismuth when transversely mag- 

 netized and cooled was also discovered in 

 the course of these experiments. The study 

 of dielectric constants at low temperatures 

 has resulted in the discovery of some inter- 

 esting facts. A fundamental deduction 

 from Maxwell's theory is that the square 

 of the refractive index of a body should be 

 the same nujnber as its dielectric constant. 

 So far, however, from this being the case 

 generally, the exceptions are far more nu- 

 merous than the coincidences. It has been 

 shown in the case of many substances, such 

 as ice and glass, that an increase in the 

 frequency of the alternating electromotive 

 force results in a reduction of the dielectric 

 constant to a value more consistent with 

 Maxwell's law. By experiments upon 

 many substances it is shown that even a 

 moderate increase of frequency brings the 

 large dielectric constant to values quite 

 near to that required by Maxwell's law. 

 It was thus shown that low temperature 

 has the same effect as high frequency in 

 annulling the abnormal dielectric values. 

 The exact measurement of the dielectric 

 constant of liquid oxygen as well as its 

 magnetic permeability, combined with the 

 optical determination of the refractive in- 

 dex, showed that liquid oxygen strictly 

 obeys Maxwell's electro-optic law even at 

 very low electric frequencies. In magnetic 

 work the result of greatest value is the 

 proof that magnetic susceptibility varies 

 inversely as the absolute temperature. 

 This shows that the magnetization of para- 

 magnetic bodies is an affair of orientation 

 of molecules, and it suggests that at the 

 absolute zero all the feebly paramagnetic 

 bodies will be strongly magnetic. The 

 diamagnetism of bismuth was found to be 

 increased at low temperatures. The mag- 

 netic moment of a steel magnet is tempo- 



