2 1 8 Historical Review of Experiments 



ing the operation just mentioned, resume that state of 

 expansion and that capacity for heat which is proper 

 to it at a given temperature, so that the effect of the 

 pressure to which it has been subjected partly disap- 

 pears again, just as a piece of metal which has been 

 hammered resumes its natural properties on being an- 

 nealed." 



In reply to these remarks, I will call to mind what 

 follows. 



ist. The discovery which I made, that no consid- 

 erable change had taken place in the specific heat of 

 the metallic dust produced by the friction, led me in 

 no way to the supposition that the heat excited in the 

 experiment could not come from the caloric set free. 

 I only found that the source of this heat was inex- 

 haustible. To explain this phenomenon, which has 

 never yet been explained, is the point now in ques- 

 tion, and I do not see how it can be explained except 

 by giving up altogether the hypothesis adopted in re- 

 gard to caloric. 



id. If we actually suppose (and it is far from having 

 been proved) that the simple pressing together of a 

 metal is sufficient to expel the caloric contained in it, 

 still the explanation of such a natural phenomenon 

 would be advanced little or none; for since the action 

 of the force which causes the pressure is continuous, the 

 condensation of the metal brought about by this force 

 would in a short time reach its maximum ; and if 

 really in this operation ever so much caloric had been 

 disengaged from the metal, still it would very soon 

 disperse. The rubbing surfaces, on the contrary, con- 

 tinue to give forth heat, and that always to the same 

 amount. 



