410 



Mr. R. A. Lehfeldt on the 



differences between that value and the observed were plotted, 

 and a parabolic curve drawn to represent them ; in this way 

 I arrived at the following equation for the refractive index : 

 / * (calc) = l-37326 (l-.*) +1-50238 a? -'00100 + 00400 (i-^) 2 . 



from which the numbers in the fifth column above are 

 derived. 



A table showing the relation between composition and 

 refractive index was then prepared, and by reference to it the 

 composition of any mixture examined with the refractometer 

 could at once be determined. 



In the same way for 



Benzene and Methyl Acetate. 



Per cent, 

 benzene. 



Temp. 



fi (obs.). 



reduced to 



18°. 



(additive). 



(calc). 



A* calc. A'obs. 







O 



19-4 



1-36283 













17-2 



(mean 



1-36293 



1-36288) 









9-38 



170 



1-37575 



1-37596 



1-37569 



-•00006 



16-88 



17-5 



1-38586 



1-38642 



1-38597 



+ -00011 



32-46 



176 



1-40769* 



1-40816 



1-40746 



- -00023 



45-16 



172 



1-42511 



1-42588 



1-42509 



- -00002 



63-76 



17-2 



1-45100 



1-45183 



1-45109 



+ -00009 



75-45 



172 



1-46739 



1-46813 



1-46754 



+ -00015 



93-74 



16-8 



1-49356 



1-49357 



1-49339 



- -00017 



100 



18-8 



1-50246 









100 



16-8 

 (mean 



1-50231 



150238) 









* Observation doubtful. 



The numbers in the fifth column are derived from the 

 equation 



/*= 1-36288 (1-tf) +1-50238 tf--000S0 + -00320(i-tf) 2 . 



After these preliminary experiments, a number of mixtures 

 were distilled as described above ; the liquid in the flask, 

 sampled before and after, showed no appreciable change in the 

 case of the ethyl mixtures, and in that of the methyl rarely 

 more than 1 per cent. The mean value was therefore taken 

 as giving q the composition of the distilland. Of the distillate 

 three samples were usually taken ; the first occasionally con- 

 tained a slightly lower percentage of benzene, owing to the 

 greater condensation of that liquid — the less volatile — on the 

 delivery tube ; the others usually agreed to within one or two 

 tenths per cent. The results are given in the 

 tables : — 





following 



