350 Mr. J. H. J. Poole on a Possible Connexion between 



a flat polished bar of iron which was placed across the poles 

 of a small electromagnet. When the magnet is excited the 

 lines of force in the central portions at least of the specimen 

 will be very approximately parallel to the surface. As it is 

 necessary to raise the specimen to a high negative voltage to 

 help the electrons to escape it must be insulated from the 

 pole pieces of the electromagnet. This was done by inserting 

 thin mica insulation between the two. This plan has also 

 the advantage of introducing an air gap in the magnetic 

 circuit, as otherwise it is possible that the residual mag- 

 netism when the field was turned off would be quite 

 large. 



Only the electronic current from a small central area of the 

 iron specimen was measured. The rest of the surface of the 

 bar was coated with a fairly thick coat of shellac which 

 would prevent any electrons escaping from the iron surface. 

 Opposite the uncoated portion of the iron a copper gauze 

 grid was placed at a distance of a few millimetres. The grid 

 was supported on sulphur insulation and connected to the 

 electrometer as before. The area of the grid was larger than 

 that of the active iron surface, and the ultra-violet light 

 passed through the grid as before. The illumination was 

 always normal to the surface. The method of procedure was 

 exactly as previously described, except that alternate readings 

 with the field off and then on were taken. 



There is a point which, however, arises in connexion with 

 the possible escape of photo-electrons from the shellac which 

 perhaps ought to be considered. H. S. Allen in his text-book 

 on 'Photo-electricity' (p. 83) states that W. Wilson found that 

 shellac was non-photo-electric, and it was for this reason that 

 shellac was adopted as a means of restricting the area from 

 which the electrons could escape. It was subsequently found, 

 however, that shellac, as a matter of fact, is slightly photo- 

 electric, about l/40th the activity of zinc, so probably a small 

 proportion of the electrons reaching the grid came from the 

 shellac. This, however, would probably not vitiate the 

 results, as presumably the number of such electrons 

 escaping would not depend on whether the iron was mag- 

 netized or not, as it has been previously shown that the 

 direct effect of the magnetic field on the path of the gaseous 

 ion is negligible. 



Table IV. gives a short summary of the results obtained. 

 It will be seen that it also gives no indication of any change 

 in the total electron current on magnetization. Each result 



