22 THE ROYAL SOCIETY OF CANADA 
For this experiment we may put 
ui — -0000188 
DA: 
d'A AMOR AU 
KS -0917 
d'=— Seas 
The substitution gives 
1/q — 1/q’ = -0031 xX 10". 
The agreement between theory and experiment leaves no such 
large gap to be filled as did the results of Wertheim; that the approach 
is so good suggests that it might be profitable to extend such experi- 
ments over a much larger range of substances. 
I have to thank Messrs. J. D. Buchanan and C. W. Robb of the 
class of 1912, for making many of the observations recorded above. 
Department of Physics, 
University of Toronto, 
May 15th, 1911. 
