PROFESSOR IN BERLIN 329 



W, Thomson had extended this theory to the motions of rigid 

 bodies in magnetizable fluids, and shown them to be related to 

 Faraday's diamagnetic experiments. So soon as the molecules 

 of magnetic or electrically polarized media can be displaced in 

 relation to each other, molecular action necessarily comes into 

 play as well as the original forces acting at a distance. Faraday 

 had assumed a state of tension in the magnetically or dielec- 

 trically polarized media in the direction of the lines of force, in 

 consequence of which these tend to shorten, while a pressure acts 

 at right angles to these lines, which tends to drive the substance 

 out in that direction. After W. Thomson in 1843 had proved 

 that forces of this nature could produce the same effect as direct 

 action at a distance on Coulomb's theory, Clerk Maxwell had 

 made this assumption of Faraday the basis of his whole theory 

 of electricity and magnetism. The remarkable effort of electric 

 insulators to expand transversely to the direction of the elec- 

 trical lines of force had already been established by experiment, 

 when Helmholtz, in his communication to the Berlin Academy 

 (Feb. 17, 1881) 'On the Forces acting on the Interior of Magnetic 

 or Dielectrically Polarized Bodies ', proposed a complete theory 

 of the phenomenon that insulators tend to alter their shape 

 under the influence of dielectric forces. 



He shows that the tensions which produce an expansion 

 perpendicular to the lines of electric force are (without any 

 special assumption as to the internal constitution of dielectric 

 media) a necessary consequence of the law of conservation of 

 energy, and of those laws which by Poisson's theory regulate 

 temporary magnetism, and are directly transferable to dielectric 

 polarization. By supposing that the constants in Poisson's 

 equations may alter, on the one hand, in consequence of the 

 altered density of the medium, on the other in virtue of the actual 

 displacement, he arrives at another distribution of the potential, 

 and thus at a calculable alteration of energy. But since the 

 equivalent of these is the excess of work which the pondero- 

 motive forces must accomplish to produce the displacements of 

 the points, beyond what is required when no dielectric tension is 

 present, he was able to calculate these forces where the change 

 of energy is determined. The discussion of how far the calcu- 

 lated forces may be resolved into molecular forces shows that it 

 is possible to replace them by a pressure that acts within a 



