on Magnetic Rotatory Polarization in Gases. A%7 



are very small, and the precision of the observations is much 

 less than for the other gases. It is noticeable that this want 

 of precision was characteristic of the experiments made with 

 oxygen — although I always took the measurements with the 

 same care, and although I studied this gas several times at 

 different periods during our series of experiments. If the 

 numbers found directly for the various colours be not cor- 

 rected by the rotation of the glasses, it will be seen that they 

 are very nearly equal to each other. In order to prove this 

 fact, I took the precaution of making a large number of series, 

 operating alternately with red and green rays; and I always 

 found that the rotations were essentially the same. 



Now, as the magnetic rotation due to the glasses is greater 

 for the green rays than for the red rays, we should conclude 

 from this that the oxygen would deflect the plane of polariza- 

 tion of the green rays less than that of the red rays, in oppo- 

 sition to the other gases studied up to this time. This body 

 presents, then, a most curious exception; and it would be very 

 important to establish this fact irrefragably. The numbers 

 that I obtained do not as yet allow of it; but I anticipate soon 

 being able to complete the first experiments required to do so. 



However, it is easy to see that oxygen, if it does not pre- 

 sent an inverse magnetic rotatory dispersion, would have at 

 least a dispersion which is very nearly nil. The present ex- 

 periments do not allow us to decide whether the rotations 

 slightly augment or decrease with the refrangibility. 



It is not uninteresting to consider the exceptional pheno- 

 menon of the magnetic properties of oxygen. 



I have shown in previous researches that very magnetic 

 bodies endowed with a negative magnetic rotatory power dis- 

 perse the planes of polarization of light according to a differ- 

 ent law from that which governs positive rotations. The 

 ratios of the negative rotations are obviously the squares of the 

 ratios which correspond to the positive magnetic rotations for 

 the same luminous rays. 



Now, it is extremely interesting to remark that the excep- 

 tion relative to oxygen coincides with the special magnetic 

 properties of this gas. At present we cannot demonstrate 

 that the coincidence which the two phenomena appear to show 

 connects them one with the other. By referring to my former 

 researches, it will be seen that it is possible to conceive the 

 existence of a mixture of two substances, the one magnetic 

 and the other diamagnetic, which would give rise to the same 

 phenomenon as oxygen. This, however, is pure hypothesis ; 

 but it is interesting to consider it, especially when" it relates 

 to a body so singular as oxygen, which, on the one hand, is 

 endowed with exceptional magnetic properties, and, on the 



