30 Mr. Donovan on Galvanometric Deflections 



and when distilled water was substituted for spring water, the needle moved 

 20°, and stood permanently at 10°. 



Here then is an important difference established between these two agents : 

 that one developed by the attrition of metals, far from decomposing chemical 

 compounds, is not even conducted by salino-aqueous liquids. It may be said, 

 in defence of the hypothesis of identity, that the cause is a difference in quan- 

 tity or intensity of the agent concerned in both. But what appears to oppose 

 this explanation is, that whatever may be the condition with regard to quantity 

 or intensity of this agent which refuses to be conducted by salt water, it is not 

 practicable to reduce the voltaic agent to the same condition. I repeated my 

 experiment, using a thin wire of zinc and an equal one of copper, in place of 

 plates previously employed, the exciting liquid being distilled water. Thus 

 the intensity and quantity were reduced to the lowest degree of both that we 

 are acquainted with, yet the agent thus brought into operation traversed the 

 forty inches of water with ease. 



Although the ring of bismuth rubbing against antimony gave a pei'maneut 

 deflection of 7.5°, which ceased when one-fortieth of an inch of water was inter- 

 posed in the circuit, I found that when these two metals were immersed in 

 spring water, at the distance of eight inches from each other, a deflection of 4° 

 was produced. Here, then, were the same metals, and the same liquid: the 

 only difference was that in one case a feeble chemical action was in operation, 

 and an agent was in consequence generated which had the faculty of being 

 conducted by water several inches in extent ; while, in the other case, an agent 

 was produced which was altogether stopped by a drop. 



To ascertain whether the interposition of so small an interval as one-fortieth 

 of an inch of water, in the circuit of what is called a thermo-electric couple, 

 would intercept the deflecting agent, I made the experiment with hot bismuth 

 m contact with cold antimony, and found that the galvanometer was not in the 

 least degree affected. This result, instead of supporting the inference which 

 I have drawn, might perhaps be used as an argument against it ; for it is known 

 that a thermo-electric battery will cause chemical decompositions. It may be 

 said that it does so because the power is accumulated ; and that if we were 

 acquainted with means of accumulating the power of the bismuth ring revolv- 

 ing against an antimony rubber, decompositions might be equally effected in 

 that manner. This, however, is but the semblance of an opposing argument : 



