266 Prof. E. Wartniann's third Memoir on Induction. 



same time setting the electro-magnet in action. The cun-ent 

 went from north to south between the non-magnetized cylin- 

 ders of the first vessel; then from north to south between 

 those of the second vessel and the axis of the electro-magnet; 

 then from west to east and from east to west, through the 

 liquid and the axes of the magnets of the third and fourth 

 vessels; then from south to north between the cylinders and 

 in the axis of the magnet of the fifth ; lastly, from south to 

 north in the last vessel between its cylinders removed from all 

 magnetic atmosphere. Weighings made with very delicate 

 balances showed that the deposits of oxide of iron on the ne- 

 gative cylinders were not. at all affected by the magnetism. 

 The mean weight of these deposits, estimated, after having 

 heated them to above 212° F., was 1'3 gramme. 



97. Those authors who have ti'eated of the question of the 

 influence of magnets, have often confounded the part which 

 they may play in chemical actions with that which they exer- 

 cise in phaenomena of molecular arrangements and of crystal- 

 lization. I think I have proved that their part is null in the 

 first case, but I do not deny its existence in the second. Every 

 new arrangement in the particles of a body must be influenced 

 by the magnet if it is accompanied by a disengagement of 

 dynamic electricity, a more frequent circumstance perhaps 

 than is generally supposed*. 



98. My experiments, which are new in their arrangement 

 and the power of the magnets employed, invalidate therefore 

 the opinion of Von Arnim f, J. W. Ritter |, Ludicke §, Masch- 



* A reservation must however be made for the modifications of mole- 

 cular equilibrium which the changes of condition occasion. Thus tlie 

 fusion of solids and the solidification of liquids do not liberate more elec- 

 tricity than the evaporation of pure water or the liquefaction of vapours 

 chemically neutral. Any one may convince himself of this by placing in 

 a horizontal glass tube a leaden cylinder, the extremities of which are con- 

 nected with those of a gooil rheometer, and by effecting the fusion of the 

 metal or its solidification by a requisite elevation or lowering of tempera- 

 ture. Another experiment consists in filling a platina crucible with melted 

 wax, in which is suspended a great quantity of plumbago. In the midst 

 of this conducting mass is immersed a platinum rod, isolated in its centre 

 bv a covering of glass, and the lower extremity of which is naked only for 

 an extremely small length. The rheometvic wire communicating on the 

 one part with the crucible, and on the other with the projecting head of 

 the rod, the wax may be melted or be allowed to congeal without any cur- 

 rent resulting. 



f Idccn zu c'mer Theorie des Magnet'wnns. Glib. Ann., tome iii. p. 59; 

 tome V. p. 394, and tome viii. p. 279. 



X Beitrage zu naherer Kcniitniss dcs Gafvanismus, tome ii. p. 55. 



^ Gilb. ./if«n.,tome ix. p. 375. This observer for the rest has since main- 

 tained an opposite opinion. See Glib. Ann., tome xi. p. 117. 



