350 Hare on some Improved Forms, S^c. 



appeared incomparably greater. The shock from the 

 Cruickshank trough was more severe. You must recol- 

 lect, that in former experiments I found that galvanic 

 plates, with their edges exposed as they are in the porce- 

 lain troughs, used by Sir Humphrey Davy, were almost 

 inefficient, when used without insulation, as the deflagra- 

 tor pairs are. This demonstrates, that an unaccountable 

 difference is producible in galvanic apparatus, by changes 

 of form or position. 



Being accustomed to associate the idea of the zinc pole, 

 in a Voltaic series, with the end terminated by zinc, and 

 the copper pole, with the end terminated by copper, T was 

 surprised to find that, in decomposing water, the oxygen 

 was attracted by the wire connected with the copper end 

 of my deflagrator, while the hydrogen went to the wire 

 connected with the zinc end. Subsequently, however, it 

 occurred to me, that, in the deflagrator, the zinc pole is 

 terminated by copper, the copper pole by zinc ; and hence 

 the apparent anomaly, that oxygen appears to be attracted 

 by copper, and hydrogen to be attracted by zinc. 



The projection from the carbon, exposed between the 

 poles, takes place at the negative pole of the pile, and 

 not at the positive pole, as you have alleged; and thus 

 your observation, that the current of igneous matter, is 

 from the copper to the zinc, is rendered consistent with 

 the Franklinian theory. 



On inspecting the diagram, No. 3, plate 5, it maybe seen, 

 that by the section, each copper case, appears like two 

 plates C c, of which those marked C, only appear to be 

 soldered to the connecting straps. Removing the portions 

 of the cases marked c, the series would be reduced within 

 the definition, above given, of series used in apparatus on 

 the principle of the Couronne des Tasses, in which the 

 zinc is positive. It would be surprising, that doubling the 

 extent, or varying the position of the copper surfaceSj 

 should cause any change of polarity. 



The observations, which are the subject of this commu- 

 nication, combined with those which you have made, of the 



