274 Prof. G. "Wiedemann on the Methods 



for determining electromotive force without, having previously 

 determined the resistance of the rotating coil. 



Let the resistance K to be investigated and a tangent com- 

 pass be included in the circuit of a constant pile, e. <j. a thermo- 

 pile. Let the ends of the resistance be connected also by a 

 second circuit, into which a delicate galvanometer and the 

 coil rotating about a vertical axis are introduced during a 

 particular phase of its motion by means of a commutator 

 revolving with the coil. The velocity of rotation is so regu- 

 lated, or the resistance in the branch containing the constant 

 pile is so altered, while the velocity of rotation of the coil 

 remains constant, that the galvanometer is not affected. Then, 

 at the moment that the coil is introduced, the induced electro- 

 motive force in it is equal to the difference of potential at the 

 ends of the conductor R caused by the thermopile. If the 

 current-intensity in it, and also in the circuit of the thermo- 

 pile and of the tangent-galvanometer, be equal to J, then 

 E = JR. The factor of reduction of the tangent-galvano- 

 meter is to be determined from its dimensions and the hori- 

 zontal component of the earth's magnetism at the place, and 

 in the same way the electromotive force induced by the rota- 

 tion of the inductor. In most cases Ave may assume the hori- 

 zontal component of the earth's magnetism to be the same at 

 both places, as in Weber's first method, which thus disappears 

 from the calculation. In this method, therefore, we have the 

 same determinations to make as in Weber's fourth method. 

 The uncertainty of contact in the commutator is without in- 

 fluence, since when the galvanometer is not affected there is 

 no current in the inductor-circuit. If we adjust the commu- 

 tator so that it makes contact with the coil at the time of 

 maximum induction, just at the moment when it varies least, 

 then, if the ratios are in other respects properly chosen, the 

 influence of self-induction in the rotating coil is reduced to a 

 minimum. A certain amount of care, however, is required 

 in order to avoid thermo-electric disturbances, resulting from 

 the heating of the points of contact; and the determination of 

 the exact phase during which the commutator makes contact 

 offers a certain amount of difficulty. 



The results so far obtained by the different methods are, 

 some of them, tolerably far apart, although the investigations 

 have been carried out with great care, if not in all points with 

 the most perfect experimental means which a final determi- 

 nation of the ohm demands. Thus the resistance of a mercurv 

 column 1 metre long of 1 square millim. section at 0° C. Mas 

 found by the first experiments of the British Association to 



