[ 97 ] 



XII. On Torsion. By Gustav Wiedemann. 

 [Concluded from p. 15.] 



15. Rotation of the Molecules. — I have already, in my 

 memoirs from the year 1858 to 1862, endeavoured, by the 

 investigation of the mutual relations between the mechanical 

 and the magnetic behaviour of bodies when they undergo 

 alterations of form, to demonstrate experimentally, besides 

 the displacement of their molecules on one another, at the 

 same time also the alterations of direction of their axes, or 

 their rotations about their centre of gravity, and to show that 

 these displacements are constantly accompanied by the last- 

 mentioned rotations which are rendered evident by alterations 

 in the magnetism of bodies (at torsions of temporary and per- 

 manent magnets), and also that, conversely, a rotation of the 

 molecules may have for its consequence a displacement of them 

 (when a current is passed through a magnet). 



If the existence of discrete molecules in bodies be admitted 

 at all, those experiments might not only prove in general the 

 existence of preformed molecules capable of rotation about 

 their centre of gravity, of displacement, and, finally, magnetic, 

 but also determine, at least in general, the nature of the 

 rotations they undergo in individual cases of external altera- 

 tion of form (for example, in the torsion of iron wires through 

 which a current passes, &c). Thus also it can be proved, as 

 already follows from Werthehn's observations, that magnetic 

 steel wires which have first been twisted and then passed into 

 the position of permanent torsion or of absolute nil exhibit 

 polarity externally, consequently that their molecules still 

 retain a sensible rotation. The hypothesis advanced in the 

 appendix to W. Weber and Kohlrausch's paper* (1866), that 

 the elastic afteraction proceeds from a rotation of the molecules 

 producing a change in the molecular forces without the centres 

 of the molecules necessarily altering their relative position, 

 would have found from the beginning a very positive support 

 by reference to these results. 



The above results, and especially the indication of the mag- 

 netism of a wire which, having been first transversally mag- 

 netized by a current passing through it, and undergone tor- 

 sion, is fully restored to the zero position, in conjunction with 

 the observations on the peculiar behaviour of bent and then 

 unbent, or twisted and then untwisted bodies under changes 



* Pogg. Ann. cxxviii. p. 414 (1866) ; also Braun, Pogg. Ann. clix. 

 p. 337 (1876). 



Phil. Mag. S. 5. Vol. 9. No. 54. Feb. 1880. H 



