Mr. I. B. Cooke on the Measurement of Chemical Affinity. 87 - 



cient resistance could not be obtained for this purpose through 

 their use, except by a most inconvenient length of wire, even of 

 the utmost practicable tenuity, and though used as an indepen- 

 dent resistance, without directive action on the needle. Fluids 

 could easily be made to oppose the required resistance, but would 

 introduce new tensions into the circuit, and perhaps additioilal 

 sources of inconstancy in their electrolysis. 



Theoretically, the electrolysis of a salt, between electrodes of 

 its metal base, should introduce only a balance of affinities into 

 the circuit ; inasmuch as for every equivalent of the metal thrown 

 down at the cathode, an equivalent is absorbed at the anode. 

 But practically, it is almost impossible to place two pieces of the 

 same metal, in even a solution of their own salt, without a cur- 

 rent of electricity being generated between them when put in 

 contact ; and if a current be passed between them from an inde- 

 pendent source, a reactionary force is generally created. These . 

 evils we might also perhaps hope to overcome by such a consi- 

 derable reduction of the quantity of the current, as would leave 

 merely sufficient to be measured by a galvanometer of the most 

 sensitive construction. 



A glass tube 4 inches long, and about yoth of an inch bore, 

 was fitted with a copper cap at each end. It was then by means 

 of a small opening drilled through each cap, filled with a solu- 

 tion of sulphate of copper, formed of one part of the saturated 

 solution with ten of water. The openings were then closed tight 

 by small copper wires screwed in. 



When this tube was connected in a vertical position with one 

 wire of a very sensitive galvanometer, the current from a pair of 

 zinc and copper plates in rain-water, exposing a surface of 1 

 square inch and | inch apart, if sent through so as to deposit 

 copper upon the lower cap, caused a deflection of the needle of 8°, 

 which was perfectly constant through long -continued action. 



The two caps of the tube afterwards showed no tendency to 

 reaction through the galvanometer. All pairs of zinc and copper 

 in rain-water, without reference to size, if only not smaller than 

 the above, gave exactly the same indication through the tube and 

 galvanometer of 8°; and in all these cases it was perfectly . 

 constant. 



If the copper plate of the electromotive pair was retained at 

 the standard size of 1 inch, the zinc plate might be considerably 

 reduced below that limit, without diminishing the maximum de- 

 flection of 8° ; but however large the zinc plate was made, the 

 copper plate could not be materially reduced below the standard 

 size of 1 inch, without reducing the deflection and rendering the 

 current inconstant. 



The resistance furnished by the tube was thus evidently in- 



