1884.] On Magnetic Polarity and Neutrality, 413 



requires that there should be a sufficient thickness in a bar of iron or 

 steel in order to produce asymmetrical opposing polarity. Coulomb's 

 theory of the neutrality taking place in the molecule itself requires 

 no thickness except that of a molecule. Ampere's theory could 

 allow of heterogeneity on the surface as easily as in the interior, con- 

 sequently thickness of a bar would, according to these theories, have- 

 no favourable result ; but if the theory that I have advanced is true,, 

 thickness should have the greatest possible influence. An extremely 

 thin strip or bar of iron should have an infinitely higher proportionate 

 remaining magnetism from the want of interior reaction, whilst an 

 extremely large solid bar should have infinitely less proportionate 

 remaining magnetism. This at once allows us to test the truth of the 

 theory by an independent method free from all experimental errors, 

 as we may place in the coil of the magnetic balance bars of iron or 

 steel of different degrees of thickness, observe their magnetic capacity 

 whilst under the influence of an electric current, and the degree of 

 remaining magnetism on its cessation, and note the extraordinary 

 influence which thickness has in allowing space for the opposing waves 

 of polarity to produce instantly a higher degree of neutrality than 

 is possible without its aid. 



The conditions of the experiments are really those of ordinary 

 electro-magnets, the iron or steel under observation is simply at the 

 time of observation a core of an electro-magnet. 



Numerous experiments were made on this subject, all confirming 

 the views advanced. A few examples will be sufficient to include 

 them all, for if we place in the coil of the balance different thicknesses 

 of the same diameter and length of iron or steel, we notice a marked 

 rise in its exterior force or magnetic capacity while under the influ- 

 ence of the electric current, and upon its cessation an equally marked 

 return to a more perfect neutrality with each increase of thickness. 

 The table on the following page contains sufficient examples to show 

 this clearly. 



This table gives the results of round cores ; experiments, how- 

 ever, were also made with flat bars with like results, the form or 

 length having no direct influence, as the reactions are transversal 

 and localised from a point in the exterior to one in the interior. 



Comparing No. 1 of the table (consisting of an extremely thin 

 sheet-iron tube) with No. 2 (a solid bar of iron of exactly similar 

 size), we have for the thin tube a remaining magnetism of 50 per 

 cent, of its previous polar force, and in the solid bar we have only 

 3 per cent. ; whilst in the solid bar, where the opposing waves of 

 polarity could easily form and produce a near approach to neutrality, 

 we find that its polar force under the influence of the coil is 400 per 

 cent, greater than that of the thin tube. 



Although, as well known, hard steel has a higher retaining power ? 



