dependent on the Motion of the Conductor, 93 



the positions of the wire of the magnetometer were made use 

 of, in which the resistance and the number of the turns were 

 equal. In the one position the ends of the wire are connected 

 with the binding-screws / and g y and in the other with h and n. 

 At K an electromotor is inserted, one pole of which is connected 

 with g and A, and the other with the gold wire in the middle 

 side-tube d. The gold wires of the outer side-tubes c and e 

 are united by means of a conducting-wire with /and n. 



If the glass tube is filled with liquid, and this is at rest, the 

 current from the electromotor divides itself between the two 

 conductions, and these two portions of the current go through 

 the turns of the magnetometer in opposite directions. The 

 resistances of the two conductions are as equal as possible ; and 

 the polarization in the one must be just in the same proportion 

 equal to that in the other, because the polarization of the gold 

 disk in the middle side-tube exerts the same action upon both 

 portions of the current, and the polarization of the two outer 

 polar disks must be of the same nature and amount. Both 

 portions must therefore be nearly the same in quantity, and 

 the sum of their actions on the magnetometer become nearly 

 equal to zero. The experiments also showed that the position 

 of equilibrium of the needle-system differed by only a few 

 scale-divisions from that occupied by the system when no cur- 

 rent was passing through the windings of the magnetometer. 



If, for instance, the liquid is moving from a to b, one portion 

 of the galvanic current goes in the same direction as the flow- 

 ing liquid, but the other portion goes in the opposite direction 

 against the liquid stream. If the positive pole of the electro- 

 motor is connected with the gold disk in the middle side-tube, 

 the stream of liquid and the portion d e of the galvanic current 

 go in the same direction, but the portion c d in the reverse di- 

 rection to that of the liquid ; but if the negative pole of the 

 series is connected with the same gold disk, the conditions are 

 reversed. Now, if the resistance is dependent on the motion of the 

 liquid (as the theory requires), the two portions of the current 

 become unequal, and consequently the needle of the magneto- 

 meter must make a deflection, the direction of which can easily 

 be determined beforehand. But there is another reason that 

 the needle of the magnetometer may alter its position of equi- 

 librium : as soon as the liquid is put in motion the polarization 

 of the three gold disks is changed. Yet the change in the 

 polarization of the middle plate has no influence upon the rela- 

 tive intensity of the two parts of the current, because this plate 

 is common to both, and a change in the polarization must act 

 with equal effect on both currents ; it therefore cannot occasion 

 a deflection of the needle. The outer plates do not behave in 



