Intelligence and Miscellaneous Articles. 451 



thermoelectric power of antimony and bismuth is destroyed by 

 removing them from all contact with oxygen and immersing them 

 in an atmosphere of pure nitrogen. From this he argues that the 

 thermoelectric force in general is due to the contact of the gases 

 -which bathe the metals. The following experiment was tried to 

 test the theory : — 



By the kindness of Mr. Edison and Mr. Upton a vacuum-tube 

 was prepared in Mr. Edison's laboratory, containing an iron wire, 

 about two inches long, firmly joined to two platinum terminals 

 which passed through the walls of the tube. The tube was ex- 

 hausted until the spark from a two-inch induction-coil would not 

 pass g of an inch in the gauge-tube, indicating a residual atmo- 

 sphere of about one-millionth. The wire was heated to incan- 

 descence during the exhaustion, in order to drive off any possible 

 occluded gases. The platinum wires outside the tubes were joined 

 to iron wires, the joinings being covered by glass tubes slipped over 

 them ; and a sensitive reflecting galvanometer was included in the 

 circuit. By laying the tube and connected joinings in the sun- 

 shine, and alternately shading one or several of the joinings, it was 

 found that the electromotive power of the joinings within the tube 

 was precisely the same as that of those without, and the develop- 

 ment of current just as rapid. There was no trace of any 

 modification due to the exhaustion. — American Journal of Science, 

 November 1880. 



ON INDUCTION IN ROTATING SPHERES. BY H. HERTZ*. 



The reciprocal actions between magnets and rotating masses of 

 metals, discovered by Arago, were first apprehended by Faraday as 

 phenomena of electrodynamic attraction, and traced to currents 

 which are induced in the masses by the magnets. 



In the present work the problem, to determine these currents 

 from the mathematical theory, is solved for the case that the body 

 considered is a solid or a hollow sphere rotating about a diameter. 

 The inducing magnets can be situated in the external or, with hol- 

 low spheres, in the internal space. The solution is also extended 

 .to the case in which the mass of the sphere is capable of assuming 

 magnetic polarity. The principal results are collected in the fol- 

 lowing summary ; for the workings I must refer the reader to the 

 original. 



As in the sphere electrical motions can be superposed, the indu- 

 ciug-potential function is resolved according to spherical functions, 

 and a term of the evolution of the form Ajo w cos i wP m (tf) considered. 

 Eor the induction produced by such a term the following proposi- 

 tions were demonstrated : — 



The flow always takes place in concentric spherical shells around 

 the zero point. The current-function of each spherical layer is a 

 spherical function of the same kind as the inducing ; the current- 

 curves have therefore the same form as the level-curves of the in- 

 ducing potential ; but the)'' appear rotated towards them a certain 



* Inaugural Dissertation : Berlin, 1880. 93 pp. Abstract by the Author. 



