dependent on tlie Motion of the Conductor. 103 



A. 



B. 



D. 



615-0 



617-0 



1-6 



615-8 



616-2 



0-8 



615-0 



616-2 



1-2 



615-0 



616-0 



1-0 



615-0 



Mean . 



. 1-15 



After this the glass tube was turned back again ; all beside 

 unaltered. The following observations were made a day sub- 

 sequent to the foregoing : — 



A. 



B. 



D. 



612-4 



610-5 



2-2 



613-0 



610-0 



3-0 



613-0 



610-0 



30 



613-0 



610-0 



2-5 



612-0 



610-0 



2-2 



612-4 



Mean . 



. 2-58 



The directions of the deflections in this series were exactly 

 the same as in the previous ones, indicating that the resistance 

 to conduction is lessened when the conductor moves in the 

 same direction as the galvanic current. 



In the experiments here communicated the direction of the 

 deflections is exclusively of importance, and too much weight 

 must not be attributed to their absolute magnitude ; for the 

 latter was altered partly by polarization. After the water is 

 set in motion, the two parts into which the galvanic current 

 divides are no longer equal : that which goes in the same di- 

 rection as the liquid has become greater, while the other is 

 diminished by an equal quantity; and therefore currents of 

 different intensities pass through the disks in the two outer 

 side-tubes. As the polarization, when the currents are as 

 feeble as they were in these experiments, increases with the 

 current-intensity, the more intense part of the current must be 

 somewhat more weakened by the polarization than the other ; 

 and since the deflections are proportional to the difference of 

 the two current-intensities, they must be smaller than when 

 no polarization takes place. On the other hand, the polariza- 

 tion of the middle disk, as already remarked, has no effect on 

 the relative quantity of the two parts of the current. There- 

 fore no reliable measure of the change produced in the resist- 

 ance of the liquid by its motion can be obtained from the ex- 

 periments made ; they show only that such a change actually 

 takes place, and that it cannot fall below the limit which can 

 be calculated from the deflections obtained. 



As already stated, we get no deflections, unless an electro- 



