Structural and Magneto-optic Rotation. 425 



lute " alcohol will answer the purpose so far as the transpar- 

 ency of the mixture is concerned, and any slight solvent effect 

 can be eliminated by preparing the mixture in a test tube 

 containing some powdered sodium chlorate, which is further 

 required as an indicator since the carbon bisulphide must be 

 added finally drop by drop till the maximum transparency is 

 secured, then filtering into the polarimeter tube containing the 

 powder to be tested. 



The greater specific gravity of the sodium chlorate necessi- 

 tates continual rotation of the polarimeter tube in order to keep 

 the powder in suspension. This was effected in a very simple 

 way by slipping corks of proper size for the polarimeter over 

 both ends of the tube and turning a slight groove in each, in 

 which ran an endless string from two wheels on an axis 

 directly above. By wrapping the tube with a small piece of 

 sheet lead, sufficient weight was given it to cause a regular and 

 smooth rotation when the wheels from which it hung were 

 rotated by a small electric motor worked with one Grove cell. 

 This simple device makes it possible, with a very little rigging, 

 to rotate a tube in any polarimeter which may be at hand. 



Landolt has pointed out that the rotation should be main- 

 tained at between 50 and 80 revolutions per minute, in order 

 to have uniform suspension without the centrifugal action. 



In all of the experiments in which we employed this method, 

 0-200 grm. of the sodium chlorate was taken after the whole 

 crop of crystals had been reduced to a uniform powder and 

 thoroughly mixed. 



Effect of the presence of optically active substances in the solvent 



medium. 



During the course of this investigation the peculiar results 

 obtained made it seem worth while to repeat the experiments 

 of Pope and Kipping* previously referred to, since the sus- 

 picion had arisen that their conclusions may have been based 

 upon insufficient data, just as it might have been inferred from 

 the results recorded in Table II of this paper, had not further 

 experiment failed to verify it, that the magnetic field deter- 

 mined the form of the crystals. 



Accordingly a five per cent solution of dextrose was pre- 

 pared as they have stated and subsequently saturated with 

 sodium chlorate. Three portions of this solution were filtered 

 and allowed at the same time to crystallize by spontaneous 

 evaporation in a place free from vibrations or any apparent 

 cause of disturbance. Curiously each of the three crops was 

 found to have an excess of dextro crystals. Pope and Kipping 



* Chem. News, lxxv, 45. 



