278 



Dr. Tyndall on the Laws of Magnetism. 

 Table VI. 



20. The deviations here exhibited are such as might be sup- 

 posed to occvu', were the doubt mentioned at the commence- 

 ment of this section well-grounded. For when the weights are 

 smallj as at the beginning of the columns, if we commence at 

 the top of the rheostat a long descent is necessary before the 

 ball yields ; and it seems reasonable to infer, that, the longer 

 the descent, the greater would be the amount of lagging mag- 

 netism present in the core at the moment of separation. Two 

 combined causes would thus operate to hold the ball ; the mag- 

 netism due to the current, and that which lingers behind the 

 current. It is evident that were the latter removed, to hold the 

 ball we must increase the former. Any mode of ex])eriment, 

 therefore, which does away with this lingering portion of the 

 force of the magnet ought to give us greater detlectious than a 



