218 



the direction of the current in the helix was changed, the recession 

 was stopped and the ends of the bar were attracted. The same 

 results were obtained when, instead of changing the direction of the 

 current in the helix, the polarity of the electro -magnetic cores was 

 reversed. When the latter were so excited that poles of the same 

 quality were presented to the ends of the bismuth bar, the repul- 

 sion of the one pole was balanced by the attraction of the other, and 

 under the influence of these opposing forces the bar stood still. 



Pursuing the argument further, a south pole and a north pole 

 were caused to act simultaneously upon each end of the bismuth 

 bar ; supposing one end of the latter to be repelled by a south pole, 

 then, on the assumption of diamagnetic polarity, the same end 

 would be attracted by a north pole ; and permitting both poles to 

 act upon it simultaneously from opposite sides, we may anticipate 

 that the force tending to turn the bar will be greater than if only a 

 single pole were used. To test this conclusion, four electro-mag- 

 netic cores were made use of; the two poles to the right of the bis- 

 muth bar were of the same name, while the two to the left of the 

 bismuth bar were of an opposite quality : with this arrangement the 

 mechanical action upon the bar was greatly augmented, and the fore- 

 going anticipation completely verified. 



The bar employed in these experiments is unusually large, but it 

 does not mark the practical limit of success. All the results obtained 

 with this bar were obtained with another solid cylinder of bismuth 

 14 inches long and 1 inch in diameter. The corresponding experi- 

 ments were made with bars of iron, and it was always found that the 

 arrangement of forces which caused the attraction of the ends of the 

 paramagnetic bar caused the repulsion of the ends of the diamagnetic 

 one ; while the disposition which caused the repulsion of the ends 

 of the paramagnetic bar produced, in the most manifest manner, 

 the attraction of the ends of the diamagnetic one. As regards the 

 abstract question of polarity, the only difference between iron and 

 bismuth is the comparatively intense action of the former. In the 

 case of a magnetic body, whose capacity for magnetization does not 

 exceed that of bismuth, and in which coercive force is equally ab- 

 sent, no proof can be adduced in favour of the polarity of the former 

 body that cannot be matched by proofs of equal value in every re- 

 spect of the polarity of the latter. 



