266 
MESSRS. THOMAS MARTJJST LOWRY AND PERCY CORLETT AUSTIN 
fectly adapted to the work we have been doing, and in some respects is even more 
convenient than the mercury arcs which are now available for experimental work. The 
importance of this development maybe shown by the statement that whereas twentyyears 
ago the sodium light was the only practical light-source for use in polarimetry, this line 
is now usually left to the last because it is more troublesome and in every way less 
satisfactory to read than the lines derived from the enclosed mercury and cadmium arcs. 
Photographic observations form an essential feature of the experiments which are 
now described. The method used was described in outline in 1908 (‘ Roy. Soc. Proc.,’ 
1908, A, vol. 81, 472-474), but has reached its highest development in connection with 
observations of the rotatory power of quartz for ultra-violet light, of which a detailed 
account will be given in the third paper of the present series. 
5. Specific Potatory Power of Tartaric Acid for Light of Different 
Wave-Lengths. 
A long series of observations was made, in order to establish standard values for the 
specific rotatory power of tartaric acid in aqueous solutions at 20° C. Similar observa¬ 
tions have been made by a number of observers using monochromatic sodium light 
(Pribram and Glucksmann, : Monatshefte,’ 1898, vol. 19, p. 136), or patches picked out 
from a continuous spectrum, with the help of a spectroscope or by means of light-filters 
(Wendell, ‘ Wied. Ann.,’ 1898, vol. 66, p. 1153 ; Winther, ‘ Zeitschr. Pliysikal. Chem.,’ 
1902, vol. 41, p. 166) ; but no values have been given hitherto for the pure light-sources, 
derived from the spectra of mercury and of cadmium, which have now become the 
common standards in polarimetric work. The values placed on record in Table I. make 
good this deficiency ; they also provide data for testing, for a considerable series of 
pure monochromatic light-sources, the linear relationship between concentration and 
rotatory power which was put forward by Biot in 1834, as well as the parabolic relation¬ 
ship used by Winther in 1902. 
In order to secure exact values for the specific rotatory power of the acid, all the 
solutions for these experiments were prepared from exactly-weighed quantities of acid 
and water. Two samples of acid were used, one of them (for which we are indebted to 
Messrs. Bennet, Lawes & Co.) being a specially pure sample, containing less than 0-05 
per cent, of ash, and practically no trace of lead. In order to avoid changes of concen¬ 
tration caused by evaporation, the precaution was taken of using solutions which had 
not been filtered ; small traces of insoluble matter were found to come almost exclu¬ 
sively from the Wedgwood mortar used to powder the crystals, and, when the further 
precaution was taken of merely crushing the crystals in an agate mortar, the amount 
of solid undissolved was negligible. The rotations of the solutions were determined in 
a polarimeter-tube 6 decimetres long, which had been calibrated to check the length of 
the column, and was maintained at a constant temperature by a rapid flow of water at 
20° C. Two, three, or even four series of readings were taken, as shown in Table I., 
with solutions prepared independently of one another, until it appeared probable that 
