250 



a compass to deviate from the magnetic meridian, by setting in rota- 

 tion under it plates of copper, zinc, lead, &c. 



To obtain more visible and regular effects, however, they found it 

 necessary to reverse the experiment, by setting in rotation a powerful 

 horse-shoe magnet, and suspending over it the various metals and 

 other substances to be examined, which were found to follow with 

 various degrees of readiness the motion of the magnet. The sub- 

 stances in which they succeeded in developing signs of magnetism 

 were, copper, zinc, silver, tin, lead, antimony, mercury, gold, bis- 

 muth, and carbon, in that peculiar metalloidal state in which it is 

 precipitated from carburetted hydrogen in gas-works. In the case 

 of mercury the rigorous absence of iron was secured. In other 

 bodies, such as sulphuric acid, resin, glass, and other non-conductors, 

 or imperfect conductors of electricity, no positive evidence of mag- 

 netism was obtained. 



The comparative intensities of action of thin bodies were next 

 numerically determined by two different methods, viz. by observing 

 the deviation of the compass over revolving plates of great size cast 

 to one pattern, and by the times of rotation of a neutralized system 

 of magnets suspended over them ; and it is curious that the two 

 methods, though they assigned the same order to the remaining 

 bodies, uniformly gave opposite results in the cases of zinc and cop- 

 per, placing them constantly above or below each other according to 

 the mode of observation employed. 



The authors next investigated the effect of solution of continuity 

 on the various metals, in the course of which M. Arago's results of 

 the diminution of effect, by division of the metallic plates used, were 

 verified ; and the further fact ascertained, that re- establishing the 

 metallic contact with other metals, restores the force either wholly 

 or in great measure, and that even when the metal used for solder- 

 ing has in itself but a very feeble magnetic power, thus affording a 

 power of magnifying weak degrees of magnetism. 



The law of diminution of the force by increase of distance is next 

 investigated. It appears to follow no constant progression according 

 to a fixed power of the distance, but to vary between the square and 

 the cube. 



The remainder of this paper is devoted to reasoning on the facts 

 detailed. The authors conceive that they may be all explained 

 without any new hypothesis in magnetism, by supposing simply that 

 time is requisite both for the development and loss of magnetism ; 

 and that different metals differ in respect not only of the time they 

 require, but in the intensity of the force ultimately developable in 

 them ; and they apply this explanation not only to their own results, 

 but to those obtained by Mr. Barlow in his paper on the rotation of 

 iron. 



