116 ROYAL SOCIETY OF CANADA 
a constant at 189° or 190°, and this at a concentration where they 
are still far from being entirely dissociated. Further, there are indica- 
tions that, in every case except that of the Ca. salt, the rotation after 
falling to 180° or 190° begins to increase again. The increase is 
slight, and in spite of the fact that it occurs in so many instances 
might be attributed to errors of reading, for a mistake of two or three 
minutes would account for the difference, and the instrument only 
allowed of reading to one minute. But in a paper published subse- 
quently to the making of these observations, Rimbach (Zeitschr fiir 
Phys. Chem., 25, 253) has examined the Na. salt to a higher degree of 
dilution than I was able to do with the instrument at my disposal. 
He finds that at about 1/,, N. the rotation begins to increase and con- 
tinues to do so till 1/,, N. is reached, when it decreases once more. 
As these solutions are more dilute than any of those upon which 
the Oudemans-Landolt law is based, the latter can scarcely be accepted 
for the present as entirely confirmed. It is my intention to examine 
both the salts of mandelic acid and those of other active substances 
in a more powerful instrument and at greater dilutions to see whether 
these abnormalities can be detected and confirmed. 
