1871.] Action of Metals and Liquids. 343 



one kind of liquid, either acid or alkaline, is employed. The electrodes in 

 this arrangement must be disposed in the order represented by the figure. 



Influence of Friction . 

 To ascertain if the friction of one of the electrodes against the liquid 

 had similar effects to those produced by the direct application of heat, I 

 employed the apparatus shown in fig. 10. The sketch does not require 

 explanation. 



Fig. 10. 



Experiment 1 . — By immersing two stout copper wires vertically in an 

 acidulated solution of cupric sulphate and rotating one of them at a speed 

 of about 5000 revolutions per minute, the rotating wire became electro- 

 positive. 



Experiment 2. — With a saturated solution of borax, the rotating wire 

 was positive. 



Experiment 3. — With a solution of cyanide of potassium, the rotating 

 wire was negative. 



Experiment 4. — With stout platinum wires in an acidulated solution of 

 cupric sulphate, the rotating wire became negative. 



Experiment 5. — With platinum wires in a solution composed of 200 

 grains of carbonate of potassium in 40 ounces of distilled water, the rota- 

 ting wire was faintly positive, and similarly in a very dilute solution. 



Experiment 6. — With two platinum disks one above the other in a 

 strong solution of carbonate of potassium, revolving the upper disk at a 

 speed of about 5000 revolutions per minute made it electro-positive. 



Experiment 7. — With an acidulated solution of cupric sulphate, the 

 revolving disk became feebly negative. 



On comparing these results with those obtained by unequal temperature, 

 we find that the directions of the currents in the two classes of cases were 

 reverse with copper in solutions of acidulated cupric sulphate and cyanide 

 of potassium, and similar in a solution of borax ; and with platinum in 

 solutions of acidulated cupric sulphate or carbonate of potassium the in- 



