292 
DR. T. M. LOWRY: NATURAL AND MAGNETIC ROTATORY DISPERSION 
These differences show clearly that the calculated curve is too flat from 5351 to 4359 
and not quite flat enough from 6708 to 5461. 
In estimating the actual errors of experiment, this limitation of the calculated 
curve must be taken into account, since this consideration serves to reduce the 
average error of the 24 readings from 0’0015 to less than O'OOOS and the average error 
of 22 out of 24 readings from O’OOll to less than 0'0005 ; the average error of 
experiment is, therefore, probably not greater than 0°'0005 per mm. But even if no 
such allowance be made, it may be seen from the table that 10 out of 24 observed 
values differ by less than O’OOl, 9 more by less than 0'002, 3 more by less than 0’003 
and one each only by 0‘004 and 0'006 from the calculated values. 
8. Conipariso7i tvith Earlier' Ohse^'vations. Optical Mass-centre of Sodium Light. 
Comparison with previous observations is almost impossible, except in the case of 
sodium light, for which the figures 
are 
SORET and 
Sarasin. 
Lowry. 
R Quartz “ IV ” 
L Quartz “ II ” 
R Quartz. 
L Quartz. 
(60 mm). 
(30 mm). 
. . 
. 21-696 
21-684 
21-6988 
21-7004 
Of . . 
. 21-724 
21-727 
21-7471 
21-7485 
Mean . 
. 21-710 
21-706 
21-7230 
21-7245 
In order to secure a wider basis for comparison and incidentally to determine the 
optical mass-centre of the light of the sodium doublet, readings were taken with a 
small piece of the Isevo-quartz, 25’902 mm. in length. The rotations were 
Mercury green 
Sodium yellow 
Batio 
Ratio for Dj . 
Ratio for . 
. 661°-47 
. 562°-82 
= 0-85086' 
= 0-84975 
= 0-85164 
Whence, taking Dj = 5896-155, Dg = 5890-186, the optical mass-centre of the doublet 
is found to be 5892-64 and the ratio of the intensities of the two lines Dj : D2 = 1 : 1-42. 
These figures agree fairly closely with the numbers 5892-5 and 1:1-6 calculated by 
Landolt from the data of Lippich and Dietrich (‘ Optical Rotating Power,’ p. 403). 
Confirmation was also obtained of the observations of Zollner and of Ebert 
(‘ Landolt,’ pp. 407—413) which showed that the optical mass-centre was shifted 
* Eollovving Landolt (‘ Optical Rotating Power,’ p. 403) the less refrangible line is distinguished 
R3 Dj. 
