14 AET. 11. — T. YAMAMOTO : THE FUSION CURVES 



the correction. As the quantity of naphthalene employed in 

 other series of determinations did not differ by more than one 

 gramme from the above, the expression was employed without 

 alteration. 



In the actual measurement of the fusion point the solution 

 was allowed to cool down about half a degree below this tem- 

 perature, and then stirred vigorously for a few seconds to start 

 the crystallisation. In this way plenty of crystalline flakes sepa- 

 rated out. This was favorable to the rapid establishment of 

 equilibrium between the solid and liquid phases. The thermo- 

 meter rose to a certain point, remained constant for a while and 

 then began to fall again. The amount of naphthalene or phenol 

 which separates out in this way can be readily calculated. When 

 the supercooling amounts to JT, the quantity of the solid formed 



is — zj in gramme molecules. This must be subtracted from the 



amount of the component originally put into the solution in 

 order to get its actual quantity at the fusion point measured. 



For the solutions in which the quantity of phenol was 

 constant and that of naphthalene variable no empirical equation 

 for heat capacity was calculated out, but graphical interpolation 

 was employed instead. 



The chief sources of error lay in the sluggishness of the 

 thermometrical indication, and in the impurity of the substances 

 employed. The former source of error was rather serious, because 

 the thermal insulation w^as not very good and the mixture was 

 losing heat during the measurements. 



Naphthalene was purified by recrystallisation from alcohol, 

 dried and distilled. The boiling point was absolutely constant. 

 The melting point was 7y.9(5°C. 



