72 Hart — Action of Light on Magnetism. 



possible to reverse it without tapping. Under these circum- 

 stances it was found to be impossible to demagnetize this 

 magnet totally by any of the ordinary methods. The magneto- 

 meter used in the previous experiments, while extremely 

 delicate, was not of the slightest use in these, and therefore it 

 was necessary to measure the time of vibration of the magnet 

 itself, with and without light polarized in different directions 

 falling on its surface. The system was almost aperiodic, the 

 magnet coming completely to rest in five half vibrations. No 

 difference in time amounting to 0*2 seconds was observable out 

 of a total of 40 seconds approximately. Owing to the unex- 

 pected magnetic behavior of the iron, it was not .thought that 

 the light polarized so as to have its vibrations perpendicular to 

 the magnetic axis of the magnet would demagnetize it com- 

 pletely, but that a slight fall from completely saturated state 

 might occur, and so the experiments were continued. The 

 magnet was kept suspended in a field of 600 to 800 C. G. S. 

 units for some time ; the field was gradually diminished to 200 

 C. G. S. units and then torsion for given deflection was meas- 

 ured with and without the application of polarized light. 

 Five complete revolutions of the torsion head caused a deflec- 

 tion of 55° of the magnet without the application of light, 

 while 4 - 8 revolutions caused the same deflection when light so 

 polarized that the vibrations were perpendicular to the mag- 

 netic axis was applied to the surface. There is a possible error 

 of 20° for each of these readings. The same experiment was 

 2'epeated with light parallel to the magnetic axis of the magnet 

 and a difference in the torsion amounting to 40° was observed. 

 The quantitative difference in the two experiments is very 

 small, but the results of a great number of the observations 

 were taken and were fairly consistent. This apparently shows 

 a difference depending on direction of the plane of polarized 

 light analogous to the demagnetizing effect of an alternating 

 current- 

 Owing to the anomalous behavior of iron films when placed 

 in a magnetic field and their strong resistance to a reversal of 

 the magnetization, it was thought advantageous to attempt the 

 experiment with a nickel film, since the resistance to demag- 

 netization would probably be much less than for iron films. 

 Although the film which was used contained nearly fifty times 

 the amount of metal of the heaviest of the iron films and was 

 exposed in a field of nearly 2000 C. G. S. units, there existed 

 no appreciable magnetic moment in the nickel and the experi- 

 ment was given up. 



From the results of this investigation the following conclu- 

 sions may be drawn : 



