CONSTITUTION AND TEMPERATURE ON MAGNETIC SUSCEPTIBILITY. 97 



Acetone, CH 3 CO CH 3 . 



This substance behaved like carbon tetrachloride. 



Acetic acid (glacial), CH 3 .COOH, and Propionic acid, CH 3 .CH 2 .COOH. 



Five experiments made on these fatty acids showed that the change of x on 

 crystallization was less than 1 per cent, in each case. 



Aromatic Substances. 



Hexamethylene, 



Three experiments showed that no change of x takes place at the fusion point 

 (6 C.). 



Chloroform, CH.C1 3 . 



The susceptibility of the crystals did not differ appreciably from that of the liqxiid. 

 A control experiment on nitrobenzene gave the abnormally large value 12 per cent. 

 (which agrees with the earlier experiments with this substance) for d%. 



(7) A RELATION BETWEEN THE MAGNETIC DOUBLE REFRACTION OF ORGANIC LIQUIDS 

 AND THE CHANGE OF MAGNETIC SUSCEPTIBILITY DUE TO CRYSTALLIZATION. 



A Comparison of Experimental Results. 



The general facts which have been described with regard to the large value of 3x 

 during the crystallization of aromatic compounds and the relatively small value which 

 has been obtained in the later experiments for substances of an aliphatic nature form 

 an interesting parallel with those relating to the magnetic double refraction of 

 aromatic and aliphatic liquids investigated by MM. COTTON and MOUTON.* The 

 parallelism is particularly striking with nitrobenzene and hexamethylene. For the 

 former the induced magnetic double refraction and the value of 3x are abnormally 

 large while with the latter both effects are inappreciable. t 



I propose to give a table showing the values of the percentage change of 

 susceptibility on crystallization (3x/x) and the magnetic double refraction (M), each 

 referred to nitrobenzene as unit, for a number of substances investigated by COTTON 



* ' Ann. de Chim. et de Phys.,' ser. VIIL, vol. 19, p. 153, 1910; and ser. VIIL, vol. 20, p. 194, 1910. 

 t ' Journ. de Phys.,' ser. V., vol. 1, p. 23, 1910. 

 VOL. CCXV. A. O 



