MecJtanical Equivalent of Heat. 



149 



.-agreement with the eleetrical equivalent would be an indicatidu of 

 .the accuracy with which the practical electrical units realise their 

 intended values. 



Principle of the Experiment. 



The method to be described,! we believe, has not, been previously 

 -applied. Bailie and Fery^ proposed to generate heat, not by stir- 

 ring Avater, but by placing a copper cylinder in a rotating mag- 

 netic field produced by polyplwise currents. Their proposal is 

 open to the objection that the lines of magnetic force would prob- 

 ably have a component of motion in tlie direction of the axis of 

 the copper cylinder. Heat would then be generated to which there 

 was no corresponding couple. We have obtained a rotating mag- 

 netic field Avith a known and fixed axis, A, by the rotation of an 

 electro-magnet. A copper cylinder is mounted (see Figs. 1 and 2) 

 .so as to be capable of rotation about a vertical axis, B, and is 



2. Comptes Rendus., Vol. CXXVI., p. 1494 (1S9S). 



Fig. 1 



S 



