134 POPULAR LECTURES AND ADDRESSES. 



made was more accurate than his old frictional 

 measurement of the mechanical equivalent of 

 the thermal unit could be. The result obtained, 

 assuming the British Association ohm to be ab- 

 solutely correct, gave the mechanical equivalent 

 as 782*2 foot-pounds, instead of 772 which he 

 had made it before, and he expressed himself 

 willing to make a new determination of it by 

 the frictional method. But now let us put our- 

 selves in the position of 1867, the date of this 

 report, with these competing determinations of the 

 ohm : the one obtained by the British Association 

 method of spinning coils, and the other by Joule's 

 electro-thermal method ; taking the dynamical 

 value of the thermal unit, as given by his frictional 

 method. Supposing that this electro-thermal 

 method was right, then what we are to infer is 

 hot that the result is the mechanical equivalent, 

 but that the British Association unit was not 

 io 9 , as it was supposed to be, but io 9 x 0*98697. 

 Thus this experiment was virtually Joule's deter- 

 mination of the resistance of the British Associa- 

 tion ohm in absolute measure. Lord Rayleigh's 



