32 MECHANICAL EQUIVALENT OF AN ELECTRIC DISCHARGE. 



sively introduced in the four connecting arcs under favourable 

 circumstances, it will be observed that the heat generated 

 at each time approaches the fourfold of that generated by a 

 single battery. This, however, will no longer take place if in 

 the four connecting arcs four thermometers are simultaneously 

 introduced, inasmuch as in this case the effect which in the 

 former experiment w^as concentrated upon a single thermometer 

 is now spread over all four. The same is derived immediately 

 from the observation already mentioned, that a bad conductor 

 introduced at any place of connexion lessens the heating at the 

 other places also. Hence when we have four thermometers in 

 the circuit at the same time, the heat observed in each must be 

 much less than that observed if only a single one be used, inas- 

 much as the three others by their resistance weaken the action of 

 the current ; the same will naturally be the case with four wires 

 raised to the melting-point. 



Let us sum up the result of all the cases which have hitherto 

 been investigated, the greater number, it must be confessed, are 

 too complicated to permit of a strict comparison with the theory; 

 so far, however, as a comparison is possible, it has always proved 

 favourable to our principal proposition, and 1 am aware of no 

 established fact which speaks against the proposition. I believe 

 therefore it may be regarded as corroborated by experience. 



[J. T.] 



