488 M. Verdet on the Optical Properties developed in 



to be measured had been feeble aud incapable of altering the 

 magnetic state of the needle ; but the powerful electro-magnets 

 necessary in my experiments would have considerably affected 

 the magnetism of the needle, and the observations would not 

 have been in any way comparable with one another. Neither 

 the tempering of the needle, nor the magnetization to saturation, 

 would have been a sufficient guarantee, since the magnetic state 

 of any needle whatever undergoes, as is well known, a temporary 

 change when it is brought near a strong magnet. 



I tried, without much success, to use the action exerted by 

 the electro-magnet on an unmagnetized body, magnetic or non- 

 magnetic. Under the influence of an electro-magnet, a bar of 

 a magnetic substance acquires a temporary magnetization ; and 

 if the body has no coercive force, it is admitted that this mag- 

 netization is proportional to the magnetic action. It follows 

 hence, that the action exerted by the electro-magnet on the bar 

 is proportional to the square of the magnetic action, and it only 

 remains to measure it by known methods, i, e. by torsion or by 

 oscillation. If the body under treatment be diamagnetic, although 

 we do not know exactly in what manner the phaenoinena occur, 

 it appears beyond doubt that a sort of polarity or temporary 

 magnetization is developed, and consequently we admit that the 

 action exerted by the electro-magnet is again proportional to the 

 square of the magnetic action. Unfortuuately in both cases the 

 law in question is only an approximative one, sufficiently exact 

 when it is required, for instance, to correct the effect due to the 

 small variations on the power of an electro-magnet, but which 

 cannot be the basis of a satisfactory process, destined to measure 

 magnetic actions which vary between rather extended limits. In 

 magnetic substances there exists always a sensible coercive force 

 incompatible with so simple a law ; in diamagnetic bodies no 

 indications of the existence of a coercive force have yet been 

 observed; but the only experiments we possess, in particular 

 those of M. Edmond JBecquerel, merely give an approximative 

 law. I could therefore do nothing with this process, although 

 I must admit that it might be useful in many cases : it might, 

 for instance, be employed in confirming the constancy of the 

 magnetic action in a given space, for this research is indepen- 

 dent of the exact form of the law which represents the action of 

 an electro-magnet on a magnetic or diamagnctie bar. In my 

 experiments, on the contrary, the knowledge of this law would 

 have been indispensable. 



I thougnt of measuring the action of an electro-magnet on a 

 small solenoid suspended between its brauches. By using the 

 bifilar mode of suspension employed by M.Wilhelm Weber in 

 his electro-dynamic researches, this method would have offered 



