accompany the changes in volume of Solid Bodies. 335 

 Charge of the lever . . . 10-242 



u. 



u'. 



u". 



37-0 



440 



74-0 



47-5 



400 



74-0 



40-5 



45-0 



72-5 



47-0 



36-0 



75-0 



40-0 



44-5 



74-5 



41-0 



43-5 





Mean . . 42-2 42*2 74'0 



Charge of the lever . . . 12*758 

 u. u'. u". 



58-0 56-0 1220 



60-0 50-0 112-0 



56-0 59-0 113-0 



57*0 550 117-0 



49-0 54-5 115-5 



53-5 117-0 



Mean . . 560 54' 7 1160 



The equality of the mean values of u and u' in each series of 

 experiments is the first fact noticed in inspecting this Table. 

 Thus the heat disengaged by contraction, and the heat absorbed 

 by elongation, are equal when the contraction and elongation are 

 accompanied by external mechanical work which is equal, but of 

 contrary sign. The proportionality of heat disengaged to the 

 charge of the lever is no less easy to confirm in the numbers 

 which precede. Finally, the proportionality of the elongations 

 to the charges is proved by another series of numbers which are 

 not given here. 



All these conclusions contain nothing very new; but it is 

 otherwise with that which may be drawn from a comparison of 

 the values of u" with the values of it ! and u. The first are always 

 higher than the second; in other terms, when the wire contracts 

 without effecting an external mechanical ivork, more heat is dis- 

 engaged than when it contracts to the same extent in raising a 



weight. It is easy to see that the expression u" ~ — may be 



taken as a measure of the excess of heat disengaged in the first 

 case over the heat disengaged in the second ; and the numbers 

 obtained by experiment show that this excess is proportional to 

 the square of the charge on the arm of the lever, or (what is the 

 same thing, since the elongation is proportional to the charge) 

 proportional to the product of the charge by the elongation, that 

 is, Co the mechanical work effected in the experiment in which the 



