Magnetic State of Iron and it's Photo- Electric Properties. 341 



inside the atom. But it is natural to suppose that there 

 would be some connexion between the direction of the 

 incident light, the plane of the orbit of the atomic electron, 

 and the magnitude of the perturbation produced. On the 

 classical electromagnetic theory of light, we would expect 

 the maximum effect when the plane of revolution was per- 

 pendicular to the light beam, and, if we were able to turn all 

 the orbits into this position, we would naturally expect to 

 get a larger photo-electric current than when the orbits are 

 arranged at random, as presumably they would be in any 

 ordinary surface. In the case of a magnetic element we 

 have assumed that these revolving electrons are also re- 

 sponsible for the molecular magnetic moment, and therefore, 

 by applying suitable magnetic fields, we can insure that the 

 plane of the revolving electrons is oriented in any direction 

 we please relative to the incident light. 



Suppose we consider a plane iron surface which is emitting 

 photo-electrons under the action of a constant source of 

 ultra-violet light, and which can be magnetized perpen- 

 dicular to the surface at will. The surface in its normal 

 un magnetized condition will presumably have the orbital 

 planes of its revolving magnetic electrons arranged entirely 

 at random, but when the magnetic field is applied these 

 planes will set themselves all approximately parallel to the 

 surface. Now this alteration in the orientation of the 

 electronic orbits might have an effect on the total number 

 of electrons emitted bv the surface, if either (a) the chance 

 of a photo-electron being emitted by the atom or (b) the 

 direction of emission of the photo-electron is affected. As 

 far as the first consideration is concerned, if we suppose that 

 the light is incident normally to the surface, we would 

 naturally expect an increase in the number of electrons 

 emitted, as all the atoms are now arranged in the most 

 favourable position for the light wave to produce perturba- 

 tions in the electronic orbits. If, however, the direction of 

 emission of a photo-electron by an atom depends also on the 

 orientation of the atom, the effect of magnetization is not 

 quite so obvious. It is natural to suppose that the photo- 

 electron would be ejected in the plane of its original orbit- 

 inside the atom. Thus when light falls on an unmaonetized 

 surface the directions of the emitted electrons would be quite 

 at random, and on the average half the emitted electrons 

 would escape from the surface. If now the surface is 

 magnetized normally, at first sight it would appear that all 

 the electrons will now be emitted parallel to the surface, and 

 therefore that very few will escape, hence leading to a large 



