376 ON ELECTRICAL CURRENTS. 



error. For this purpose, I repeated the experi- 

 ment, taking however the precaution to keep 

 the liquid constantly agitated. The specific heat 

 now carae out 0.406. The cause of the smallness 

 of the result became thus apparent. The oil 

 could not carry off the heat from the wire as 

 quickly as the water, and hence the wire which 

 was immersed in the oil, became highly heated, 

 occasioning an increase in its resistance, and a 

 proportional increase in the quantity of heat 

 evolved by it. This was easily proved, by 

 placing the finger in contact with the wire, which 

 could not be retained in that position longer than 

 one or two seconds. 



The object of this communication is therefore 

 to guard the experimenter against employing 

 wires of so small a surface, as those recommended 

 in my paper on specific heat, whenever powerful 

 currents are employed ; especially when, at the 

 same time, the specific heat of a viscous liquid 

 of bad conductive power, and small capacity for 

 heat, is sought. In such cases, a large strip of 

 platinum foil w-ould be preferable to a wire, on 

 account of the extensive surface which would 

 thus be presented to the liquid. 



