Journal oj Agricultural Research 
Vol. XXX, No. 3 
254 
addition of petroleum ether would give 
a purer product each time a precipita¬ 
tion was undertaken, but the spectro- 
photometric data show that this was 
not the case, for evidently oxidized 
xanthophyll precipitated along with the 
pure xanthophyll. 
The results for experiment No. 7 
were obtained by carefully recrystalliz¬ 
ing a sample of xanthophyll five or six 
times from methyl alcohol and then 
precipitating it from chloroform by the 
addition of petroleum ether. The 
xanthophyll used was obtained from 
cowpea leaves and was not allowed to 
stand for any length of time during 
the purification process. The results 
in experiment No. 8 were obtained in a 
similar manner from xanthophyll which 
had just been'extracted from the leaves. 
Data obtained in experiment No. 9 
were gotten by a reprecipitation from 
chloroform and petroleum ether of the 
sample used in experiment No. 8. 
The xanthophyll of experiment No. 7 
and of experiment No. 8 gave the best 
speetrophotometric results and had a 
melting point of 174° C. Conse¬ 
quently, an average of the readings for 
these two samples has been taken as a 
basis for quantitative spectrophoto- 
metric determinations. 
, Table I .—The transmittancy of xantho - 
phyll in ether ° 
* Experiment No. 
Transmit¬ 
tancy 6 
Milligrams 
of xantho¬ 
phyll per 
liter 
1____ 
0.0046 
.6.72 
.0640 
3.36 
.0410 
1.68 
.4740 
.84‘ 
2...-. 
.0254 
4.20 
.0593 
3.15 
.1520 
2.10 
3..-... 
.0215 
*4.20 
.0216 
4.20 
.1380 
2.10 
1 
.1410 
2.10 
4.... 
.0232 
4.20 
5... 
.0195 
4.20 
.0198 
4.20 
6.-.— 
.0216 
4.20 
.0219 
4.20 
.1490 
2.10 
.1490 
2.10 
.. 
.0180 
4.20 
.0181 
4.20 
.1350 
2.10 
8_____ 
. 0167 
4.20 
9.... 
.0107 
4.20 
• The xanthophyll was dissolved in ether or ab¬ 
solute alcohol (0.042 g. per liter) and then just before 
the readings were made was diluted with U. S. P. 
ether. 
b For 2-centimeter cell. 
• Duplicate readings are checks on the same solu¬ 
tion.* 
In determining the purity of a 
sample of xanthophyll, measurements 
made on the spectrophotometer are 
apparently more valuable than melting- 
point determinations. Melting points 
were always taken but often these 
would be apparently correct (173°- 
174°) and yet-the speetrophotometric 
results would show a difference in the 
absorption of the different solutions. 
The solution which gives the greatest 
light absorption is here considered as 
the purest. It is quite easy to remove 
all impurities other than oxidation 
products from the material. Oxidation 
products absorb less light-than pure 
xanthophyll, and consequently a prod¬ 
uct which contains any oxidized xan¬ 
thophyll would show less light absorp¬ 
tion. The graphs in Figure 1 indicate 
the presence of oxidation products in 
some of the samples. 
In this work the xanthophyll solu¬ 
tion (0.42 gm. per liter) which absorbs 
the most light is considered as being • 
the one which most closely approxi¬ 
mates 100 per cent pure xanthophyll, 
and all of the quantitative |data have 
been based on the graph obtained from 
the transmittancy of such a solution. 
The composite curve of solutions 
(the average of experiment No. 7 and 
experiment No. 8) of xanthophyll 
which gave maximum absorption is 
plotted in Figure 2 in order to make 
the data more convenient for use and 
to avoid confusion with the other 
experimental data in Figure 1. The 
manner of using Figure 2 is the same 
as described in the paper on carotin. 5 
THE COLORIMETRIC METHOD 
The colorimetric method of deter¬ 
mining the amount of xanthophyll in 
solutions is similar to that described 
for carotin. Pure xanthophyll was 
prepared and 0.042 gram was dis¬ 
solved in a liter of ether. Colori¬ 
metric determinations were made at 
once on various dilutions of this 
solution. Three different weighings 
and determinations (Table II) were 
made on as many different days, using 
Lovibond slides and a Duboscq 
colorimeter. 
The graphs in Figure 3, which are, 
the average results of three determina¬ 
tions, show clearly how the colorimeter 
readings obtained from Lovibond slides 
5, 10, and 20 yellow, vary with the 
concentration of the ether solution 
of xanthophyll. The curves shown 
here are used in the quantitative work 
upon which the interpretation of all 
8 SCHERTZ, F. M. THE QUANTITATIVE DETERMINATION OF CAROTIN BY MEANS OF THE SPECTROPHO¬ 
TOMETER ANb THE COLORIMETER. Jour. Agr. Research 26: 383-400, illus. 1924. 
