414 



PROCEEDINGS OF THE AMERICAN ACADEMY 



In this table, column second gives the interferential equivalent of 

 the liquid, the formula of which is given in column first. Column Aj 

 gives the differences of the interferential equivalents for the constant 

 chemical difference G H^ ; the liquids of each grouja being homologous. 

 It will be at once seen that Ai is not constant in any one group, but 

 increases with the molecular weight of the liquid. From this it follows 

 that the interferential equivalents of either carbon or hydrogen, or of 

 both, are variable. The last case is most probable. The mean of the 

 differences Ai is the same for all these groups. From the above it is 

 easy to see that in the strict sense neither carbon nor hydrogen can be 

 said to possess a constant interferential equivalent. As it seemed, 

 however, practicable in this case, as in the case of the refractive equiva- 

 lents, to deduce, at least, an available rule for computing approximately 

 the interferential equivalent of a compound from the equivalents of 

 its constituent atoms, I determined the equivalents of carbon, hydrogen, 

 and oxygen by means of the data given in Table V. and obtained the 

 following values : — 



Carbon 41.46 



Hydrogen 11.84 



Oxygen 27.28 



With these values I computed the numbers of column G in Table V. 

 and of the fourth column in Table VI. 



TABLE VL 



In these tables, column Ao gives the differences between the observed 

 and computed values in the case of each liquid, and column A'}^, the 

 same differences in percentages of the observed values. The means of 

 these last are also given without reference to sign. With the aid of 

 the equivalents of carbon and hydrogen, I determined those of chlorine, 

 bromine, and iodine, given in Table VIII. Finally, the equivalents of 



