l^ogal Jn^titution of OSreat IS 



WEEKLY EVENING MEETING 

 Friday, February 29, 1884. 



SiE William Bowman, Bart. LL.D. F.E.S. Honorary Secretary and 

 Vice-President, in the Chair. 



Professor D. E. Hughes, F.E.S. M.B.I. 



Theory of Magnetism. 



The theory of magnetism, which I propose demonstrating this evening, 

 may be termed the mechanical theory of magnetism, and, like the 

 now well-established mechanical theory of heat, replaces the assumed 

 magnetic fluids and elementary electric currents by a simple, sym- 

 metrical, mechanical motion of the molecules of matter and ether. 



That magnetism is of a molecular nature has long been accepted, 

 for it is evident that, no matter how much we divide a magnet, we 

 still have its two poles in each separate portion, consequently we can 

 easily imagine this division carried so far that we should at last 

 arrive at the molecule itself possessing its two distinctive poles, con- 

 sequently all theories of magnetism attempt some explanation of the 

 cause of this molecular polarity, and the reason for apparent neutrality 

 in a mass of iron. 



Coulomb and Poisson assume that each molecule is a sphere 

 containing two distinct magnetic fluids, which in the state of neutrality 

 are mixed together, but when polarised are separated from each other 

 at opposite sides ; and, in order to explain why these fluids are kept 

 apart as in a permanent magnet, they had to assume, again, that each 

 molecule contained a peculiar coercive force, whose functions were to 

 prevent any change or mixing of these fluids when separated. 



There is not one experimental evidence to prove the truth of this 

 assumption ; and as regards coercive force, we have direct experi- 

 mental proof opposing this view, as we know that molecular rigidity 

 or hardness, as in tempered steel, and molecular freedom or softness, 

 as in soft iron, fulfil all the conditions of this assumed coercive force. 

 Ampere's theory, based upon the analogy of electric currents, 

 supposes elementary currents flowing around each molecule, and that 

 in the neutral state these molecules are arranged hap-hazard in all 

 directions, but that magnetisation consists in arranging them sym- 

 metrically. 



The objections to Ampere's theory are numerous. 1st. We have 



no knowledge or experimental proof of any elementary electric 



currents continually flowing without any expenditure of energy. 



2nd. If we admit the assumption of electric currents around each 



Vol. XI. (No. 78.) b 



