ON PAKAMAG-NETIC AND DiAMAGNEttC FOECES. 32? 



diamagnetic polarity, the action of the bismuth ought 

 to exceed that of the antimony. Experiment proves 

 this to be the case. Hence the deflection produced by 

 these metals is due to their diamagnetic, and not to 

 their conductive capacity. Copper cylinders were next 

 examined : here we have a metal which conducts elec- 

 tricity fifty times better than bismuth, but its diamag- 

 netic power is nearly null; if the effects be due to 

 induced currents we ought to have them here in an 

 enormously exaggerated degree, but no sensible deflec^- 

 tion was produced by the two cylinders of copper. 



It has also been proposed by the opponents of dia- 

 magnetic polarity to coat fragments of bismuth with 

 some insulating substance, so as to render the formation 

 of induced currents impossible, and to test the question 

 with cylinders of these fragments. This requirement 

 was also fulfilled. It is only necessary to reduce the 

 bismuth to powder and expose it for a short time to the 

 air to cause the particles to become so far oxidised as 

 to render them perfectly insulating. The insulating 

 power of the powder was exhibited experimentally; 

 nevertheless, this powder, enclosed in glass tubes, ex- 

 hibited an action scarcely less powerful than that of the 

 massive bismuth cylinders. 



But the most rigid proof, a proof admitted to be 

 conclusive by those who have denied the antithesis of 

 magnetism and diamagnetism, remains to be stated. 

 Prisms of the same heavy glass as that with which the 

 diamagnetic force was discovered, were substituted for 

 the metallic cylinders, and their action upon the mag- 

 net was proved to be precisely the same in kind as that 

 of the cylinders of bismuth. The enquiry was also 

 extended to other insulators : to phosphorus, sulphur, 

 nitre, calcareous spar, statuary marble, with the same 

 invariable result : each of these substances was proved 

 22. 



