Dec. i, 1933 
Quantitative Determination of Carotin 
399 
after the first readings were made. All agreed that matching the two 
solutions was very difficult. 
Table XI .—Readings made by different observers on the same solution, showing the 
inaccuracies of the colorimeter method as generally used 
t 
1 A - 
B. 
c. 
AA. 
BB. 
47-5 
40.6 
42.5 
46.2 
43 - 6 
46.8 
43*8 
44*3 
44.6 
49.4 
47.0 
46.8 
45-4 
46.6 
45 * 8 
58.0 
59-8 
60.7 
5°-3 
60.5 
48.4 
49.8 
K2.0 
48.2 
49.4 
44.1 
45-8 
46.3 
59-9 
49.6 
When readings such as the above are obtained (Table XI) and in view of 
the unknown errors of other workers, it is readily seen that the work of 
various investigators on the carotin content of different substances can 
not be compared as to the absolute amount of carotin present. The 
results of any one worker should be of some value in giving him com¬ 
parative figures to be used only in his own work; but one does not see 
how it is possible to assign actual values to data submitted in the litera¬ 
ture and have such values mean more than a rough approximation of 
the amount of carotin actually present. 
SUMMARY 
The spectrophotometric method for the determination of carotin is 
described and graphs are given for determining the amount of carotin in 
solutions when the transmittancy is known. 
The mercury line 435.8 was finally adopted as the best scale on the 
spectrophotometer for giving the most accurate results with carotin 
solutions. 
Solutions of the same concentration of carotin in alcohol and petroleum 
ether gave approximately the same transmittancies, while the trans¬ 
mittancy for ether was a little less. 
The specific transmissive index (extinction coefficient) of carotin in 
alcohol and petroleum ether for the mercury line 435.8 was found to be 
1.91; in ether 1.986. 
Quantitative determinations of carotin by means of the colorimeter 
and the spectrophotometer were compared; the colorimeter was found 
to be unsatisfactory because of the difficulty in matching the solutions, 
thus giving a greater percentage of error, while the spectrophotometer 
gave excellent results both as to accuracy and ease of reading the 
instrument. 
In working with the spectrophotometer it is not necessary to prepare 
pure carotin as a standard or use any other standard, for the instrument 
itself is standardized in regard to carotin when the transmittancy of its 
solution is known. 
The transmittancies as measured depend upon the physical properties 
of the substances involved, and not upon variability in light, physiological 
factors, or tint of solution, which so greatly affect the readings in any 
colorimeter, and are independent of abnormalities of the observer’s color 
vision. 
71686—24 - 2 
