Hydro-Electtical Currents. 97 



be obtained with solutions in the decomposing cell which are found in 

 the waters of the ocean. Of these two, nickel, I could not procure in a 

 form to admit of high pressures being well tested ; for the other, iron. 

 In a strong voltameter filled with a saturated solution of common salt, 

 and poles of this metal, the decomposition of water proceeds readily by 

 the electro-motive force of one hydro-couple giving off hydrogen at the 

 negative, and a large dusky green insoluble deposit at the positive pole ; 

 after three or four days' slow action, the bulky deposit assumes a clearer 

 but dark hue, and eventually the pressure from the hydrogen bursts the 

 vessel. 



The currents for these decompositions were all of them derived from 

 spring or rain water acting on 1, 2, 3, or 4 pairs of small zinc and copper 

 plates ; the number of the plates being varied to supply the electrical 

 force required for the poles of the metal under examination. 



The Metah exposed to Thermo-Electric action. 



29. While engaged decomposing metallic salts by a thermo-electric 

 couple of copper and antimony fused together to form a joint, and ex- 

 cited by enveloping the joints in a blue gas-flame, I was struck by the> 

 rapid destruction of those joints, few of them enduring 48 hours. As 

 both of the metals were much oxydized, I at first attributed to the oxi- 

 dation the change noticed. But the same appearance presented itself 

 in the battery employed for the experiments described (9) ; there the 

 joints are not in the flame, and as the parts of the bars around them are 

 only coated with a thin film of oxide, corresponding in appearance to a 

 full straw colour on steel, the disruption of the joint which separated 

 with a slight touch, could not be assigned to the oxydation of the metals^ 

 From further investigations of this change in the fracture, I found it a 

 constant accompaniment of the unequal heating of thermo-electric bars 

 at temperatures above 300° ; but it is so very slow, that forty days of the 

 most energetic action of antimony and bismuth couples is necessary to 

 render the first trace of a dull fracture distinctly visible ; and with bars 

 of other metals less adapted for developing thermo-electricity than anti-^ 

 mony and bismuth, the disruption is not so rapid, the temperature of the 

 joints being equal. 



At low temperatures, it would be a valuable experiment which could 

 establish this change in the joint. So far as those I have instituted for 

 this purpose have gone, there is not a trace of the action ; the joints of 

 a battery, fig. 1, weathered for a year, are as sound as new ones. But 

 if the thermo- current given during 40 days' action of the battery, fig. 2 

 (9), must be developed, by a weathered battery, to effect decompositiona 

 to the same extent as in the first, then an estimate founded on this view, 

 with the experiments (11, 13, 19) taken as data, shews a period of from 

 40 to 50 years to be required to produce the change sought for. 



30. When the joints were found to be subject to a gradual change, an 

 experiment was made to ascertain if a long-used joint lost any of ita 



VOL. XXXVI. NO. LXXI. JAN. 1844. G 



