RECENT ADVANCES IN SCIENCE 443 



liquids the opposition to unidirectional pointing is to be sought 

 in the gyroscopic resistance of the rapidly rotating molecules 

 to change of direction, which thus endows them with a quasi- 

 rigidity, making each molecule behave very much as if its axis 

 of rotation were acted on by a restoring couple proportional to 

 the angle through which it is displaced and proportional also 

 to the kinetic energy of rotation. This sort of rigidity is well 

 exemplified in the well-known experiment in which a smart 

 blow sets quivering the inner ring of a gyroscope whose flywheel 

 is turning rapidly. 



The polarisation of the substance thus has two extremes ; 

 it is zero when the axes of the doublets are uniformly dis- 

 tributed ; it has a maximum value when they all point in one 

 direction. The greater the electric force the more highly 

 polarised is the medium for a definite energy of rotation in the 

 molecules, i.e. for a definite temperature ; for a definite electric 

 force the less polarised is the medium the greater the energy of 

 rotation of the molecules, i.e. the higher the temperature. 

 Thomson develops a formula connecting polarisation with 

 electric force and molecular rotational energy. He expresses 

 it by a graph in which polarisation is the ordinate and the ratio 

 of electric force to the mean energy of the molecule is the 

 abscissa. The graph rises from the origin asymptotically to the 

 limiting value of the polarisation, and is practically straight in 

 its initial sloping part, and then begins to bend over towards 

 the axis of abscissae, ultimately merging into a line parallel to 

 this axis. 



For a finite value of the polarisation we will have a number 

 of chains of polarised molecules all pointing in the direction of 

 the electric force, the remaining molecules being uniformly 

 distributed. Now if we consider two adjacent molecules in a 

 chain, presenting oppositely charged ends to one another, there 

 will be a considerable attraction due to the positive end of one 

 molecule, on the electrons in the negative end of the other, and 

 Thomson's essential hypothesis at this point of the argument is 

 that in the molecules of metals, in contradistinction to the 

 molecules of insulators, there exist electrons (not probably 

 those of the doublets) easily abstracted if the molecules are 

 crowded together. So we suppose that under these " end " 

 forces, electrons pass along a chain of molecules " like a com- 

 pany in single file passing over a series of stepping-stones." 



