410 PROCEEDINGS OF THE AMERICAN ACADEMY 



Line. 10% sol. 



C .' 1.34568 



G 1.35541 



Densities .... 1.03812 



/ 287.4 



1' 295.4 



lu this table, / gives the interferential constants for the three solu- 

 tions, F the constant for the liquid sugar in each, the mean value 

 beinof 297.2. As the interferential constant for water is 28G.2, it is 

 easy to see that nothing is to be hoped for from the employment of the 

 interferential constant as a means of determining the quantity cf sugar 

 in a solution, since it is clear that the degree of accuracy to be attained 

 by the method above given, will, in general terms, be in proportion to 

 the difference between the interferential constants of the constituents 

 of the given mixture. In its application to quantitative determinations, 

 the new 02)tical method is analogous to the well-known process of 

 indirect analysis, the success of which depends upon the difference 



between the atomic weights of the bodies sought. Landolt has shown 



n 1 



that tlie function — -^ — is so nearly constant for a given ray and given 



substance, that for chemical purposes no very sensible error is made in 

 considering it as absolutely constant. He has further shown that in 

 the case of a mixture of two substances we have very nearly 



This expression may be employed for the analysis of mixtures, and 

 in many cases leads to valuable results, as Landolt has sufficiently 

 shown. I am disposed to think that the method which I have pro- 

 posed above will enable us to obtain a still greater degree of accuracy 

 in cases in which the values of the interferential constants have been 

 determined with the requisite precision. 



The valuable data furnished by Wullner are not the only ones which 

 I have discussed. * Landolt and Ilaagen have also given a series of 

 measurements of the densities and indices of refraction of a number of 

 liquids. Their results are contained in Table IV. ; I have arranged 

 them for convenience in six groups, the sixth group containing the 

 data furnished by Haagen.f 



* Pogg. Annalen, T. 122, p. 545. 

 t Pogg. Annalen, T. 131, p. 117. 



