282 MAJOR A. E. OXLEY ON THE INFLUENCE OF MOLECULAR 



molecule possesses no resultant magnetic moment. For example, while iron below the 

 critical temperature possesses, according to WEISS, 11 magnetons to the atom, the 

 molecule of ferric sulphate possesses 30, ferric chloride 28, sodium ferro-pyrophosphate 

 26, sodium ferrous oxalate 27.* Again, nickel below the critical temperature possesses 

 3 magnetons to the atom, above the critical temperature either 8 or 9 magnetons to 

 the atom, while a molecule of nickel sulphate contains 1 6 magnetons. In general the 

 mimber of magnetons, per molecule of a salt of a ferro-magnetic element, is large 

 compared with the number of magnetons associated with an atom of the pure 

 ferro-magnetic. The fact that cupric salts are paramagnetic while cuprous ones are 

 diamagnetic is interesting from our point of view. Although copper is diamagnetic, 

 yet a molecule, of cupric sulphate contains 10 magnetons. It seems as if the large 

 local atomic fields, which have been recognised in diamagnetic and ferro-magnetic 

 molecules, have the power, when the molecules approach so that their fields overlap, 

 to upset the equilibrium of the atoms in combination and redistribute their magnetic 

 elements. This is easily possible when the great intensity of the local molecular 

 field is borne in mind. In most cases a diamagnetic molecule, on account of its 

 symmetry, would, under the influence of such a field, remain diamagnetic, but each 

 orbit would lie distorted by the field and the susceptibility of the substance would be 

 slightly modified. 



(7) ON SOME ANOMALIES IN THE MAGNETIC ROTATION EFFECT. 



Diamagnetic media are in general dextro-gyric. The only exception is titanium 

 chloride which is Isevo-gyric. t Paramagnetic media are sometimes dextro-gyric and 

 sometimes Irevo-gyric, while the ferro-magnetic elements, iron, nickel and cobalt are 

 all dextro-gyric. At present no theory seems capable of accounting for these 

 anomalies and it is therefore interesting to examine to what extent the local molecular 

 field may cause the effects observed. At one time it was thought that the direction 

 of rotation probably depended merely on the diamagnetic or paramagnetic property 

 of the molecule, but experiment soon disproved this generalisation. VoiGT J suggested 

 that the production of an intense reverse field, when the external field was applied, 

 would account for the effects, but no physical explanation of a possible origin for this 

 intense reverse field was given. 



On the views of diamagnetic and paramagnetic polarity developed in these 

 researches, the necessary fields demanded by VoiGT are found in the immediate 

 neighbourhood of the molecular boundary. In diamagnetic liquids the molecules have 

 zero magnetic moments and their axes are distributed at random. The application of 



* 'Journal de Phys.,' vol. 1, ser. v., p. 974, 1911. 



t For solutions of salts in water it should be noted that VERDET'S constant for the solute is to he 

 regarded as negative if VERDET'S constant for the solvent is less than 0'0130, which is the value of the 

 constant for pure water at 20 C. 



I See P. ZEEMAN, ' Researches in Magneto Optics," p. 185. 



