Origin of Molecular Attraction. 661 



-a natural collection of molecules and the dielectric in an 

 electric field. The main one is that in the electric field there 

 is a resultant direction of electric force, whereas in a collection 

 of molecules there is none, the electric force inside a mole- 

 cule being reversed between the opposite poles of two 

 molecules. The very definite conception which reduces the 

 polarity of a molecule to two point electrons at the ends of a 

 diameter of the molecule's supposedly spherical surface, 

 becomes almost a hindrance in the case of molecules in con- 

 tact at absolute zero because of the tendency of two oppositely 

 charged point electrons to come nearly into coincidence and 

 nearly to destroy one another's effect. In the facts of 

 Nature there is no indication of any tendency to such 

 destruction of effect. We shall do better then to regard the 

 molecule as polarized throughout its mass, just in the same 

 way as the Earth is magnetically polarized, and can be re- 

 placed by an infinitely short magnet at its centre, if only 

 the magnetic moment of this is equal to that of the Earth. 

 Just as we should encounter difficulties if we treated the 

 Earth as a sphere having magnetic poles at the ends of a 

 diameter, we are confronted with similar ones if we treat 

 the molecule as an electric bipole. 



With bipoles in which the axis between the poles is a 

 diameter or nearly a diameter, the law of the inverse fourth 

 power would break down completely, as it holds only for 

 bipoles in which the axis is small compared with the distance 

 between the attracting bipoles. But if the molecule is taken 

 to be a uniformly electrically polarized sphere, which may 

 be called a uniformly electrized sphere, then its external 

 effect is the same as that of an infinitely short bipole at its 

 centre, the bipole having the same electric moment as the 

 sphere. With uniformly electrized spheres having their axes 

 similarly directed along the line joining their centres, the 

 law of the inverse fourth power becomes exactly correct at 

 all distances. I hope to show in a separate communication 

 that the conception of a uniformly electrized molecule is 

 helpful in studying the vibrations causing spectra. 



As I have shown (Phil. Mag. [6] iv. p. 606) that for many 

 molecules the electric moment e(m/p ) 1 ' 3 is actually nearly 

 equal to the product of electron charge and molecular diameter, 

 the uniform electrization of the molecule is intimately related 

 with the electron charge e of electrolysis. When a number 

 of uniformly electrized spheres are brought into contact at 

 absolute zero, we get a mass which is uniformly electrized, 

 but with the axis of electrization changing in direction so 

 that the average electric force throughout the mass is nil. 

 To get a schematic representation of such a state of affairs, 



