﻿Magnetic Rotatory Dispersion of Paramagnetic Solutions. 915 



for the solutions and for pure water*. When the water of 

 crystallization of the salt is taken into account, the mass of 

 water per c.c. is practically the same in the solution as it is 

 in the case of water alone ; the difference in the ordinates 

 therefore represents the rotation produced by the salt alone. 

 As small differences of rotation were to be measured it 

 was necessary to work with fairly concentrated solutions, 

 and the graphs refer to the following concentrations : — 



Nickel Chloride '216 molar. 



Cobalt „ . -625 „ 



Nickel Sulphate *624 „ 



Cobalt „ -594 „ 



Ferrous „ -312 „ 



Nickel Acetate -664 „ 



Cobalt „ -625 „ 



The results indicate that the cobalt atom, like the iron 

 atom, is capable of producing a negative rotation, but of 

 smaller numerical value. The disappearance of this on the 

 low-frequency side of the band may be attributable to the 

 preponderating positive effect of the (S0 4 ), (C 2 H 3 2 )} or 

 CI ions. This view is supported by the fact that when a 

 highly dispersive element (e. g., chlorine) is present the 

 residual negative effect in the ultraviolet is smaller, but the 

 point requires further investigation. In the case of nickel 

 the rotation remains positive throughout the range mea- 

 sured, but does not undergo the large increase in the 

 ultraviolet which is observed with diamagnetic substances. 

 It is possible therefore that the nickel atom, whilst not 

 able to overbalance the positive effect of the other atoms in 

 the salt, still exerts sufficient depression nearly to neutralize 

 the dispersion. From this point of view, our results indicate 

 that in respect of paramagnetism the cobalt atom occupies 

 a position intermediate between those of iron and nickel. 



The experiments are to be continued, and we hope to 

 publish results for other paramagnetics shortly. 



We beo- to tender our thanks to Prof. Wilberforce for the 

 facilities and apparatus placed at our disposal and to 

 Dr. Smeath Thomas who kindly prepared the ferrous sulphate 

 solution. 



The George Holt Physics Laboratory, 

 University of Liverpool. 



* To obtain the rotation of the salts in radians per cm. -gauss the 

 graph-readings must be multiplied by 6 , 616xl0~\ 



