Journal of Agricultural Research 
Vol. XXV, No. 3 
136 
of oxygen has been taken up, but with the beginning of oxygen absorp¬ 
tion tiie solution begins to show color, and the color gradually darkens 
as the process is continued, passing through a series of tints and shades, 
from very pale brownish orange through more intense browns that 
become redder and finally somewhat purplish as the final color is 
approached. After much oxygen has been absorbed the solution is 
nearly opaque and almost black. 
A rather prolonged study of the possibilities of using the rate of change 
in the color of the indicator solution just described, as an indication of 
oxygen absorption, led to the colorimetric method here used. An 
arrangement was finally devised by which the indicator solution may be 
examined and compared with standard color solution from time to time, 
as it gradually darkens during the period of the experiment. One or 
two standard color solutions, in bottles similar to the indicator bottle, 
are placed beside the latter for this comparison, and the examination 
is always made with transmitted light from a 60-watt Mazda incan¬ 
descent electric lamp. The light readies the bottles through a diffusing 
screen of white paper and the bottles and lamp are inclosed so that all 
light reaching the eye comes directly through the solutions to be com¬ 
pared. As the apparatus is usually operated, three bright windows are 
visible, each about 5 mm. high and 2 cm. wide, arranged in a horizontal 
row. The light of the middle window is from the indicator bottle, while 
that of either of the other two windows is from the corresponding color 
standard. By this arrangement it is not difficult to determine with 
great accuracy whether the indicator solution appears darker or lighter 
than any standard solution with which it is compared, and this deter¬ 
mination can be made almost instantly at any time during an experiment. 
After many unsuccessful attempts along other lines, Amy’s plan 7 for 
preparing standard color solutions was tried and found to be adequate 
for the present purpose. Three solutions were prepared, a red one of 
cobalt chlorid, a yellow one of ferric chlorid, and a blue one of copper 
chlorid. By mixing these three solutions in proper proportions and with 
proper dilutions (as determined empirically) a number of permanently 
colored mixtures were seemed, each one of which represents, with a high 
degree of accuracy, one of the colors traversed by the pyrogallol indicator 
solution during the early stages of its oxygen absorption. In the final 
experiments only two color standards were used. Standard A is very 
pale, and its color matches that of the indicator solution when a very 
little oxygen has been absorbed. Standard B is much darker, and was 
so chosen that its color matches that of the indicator solution at the end 
of an experiment period of convenient length. The compositions of the 
two standard solutions thus far used are shown in Table I. A 1 per cent 
solution of hydrochloric acid is used instead of water. 
Tabl,$ I .—Composition of two standard color solutions 
Basic compound. 
Pale standard 
solution (A), 
in 20.5 cc. 
Dark standard 
solution (B),J 
in 20 cc. 
Cobalt chlorid crystals. 
Gm. 
O. 07 
o- 54 
O. 144 
Gm. 
I. OO 
O. 14 
O. 20 
Ferric chlorid crystals. 
Cupric chlorid crystals. 
7 Arny, H. V., and Ring, C. H. standardized colored fluids. In Jour. Franklin Inst., v. i8o, p. 
199-313. 1915. 
