May 3,1924 Modifications of the Picric Acid Method for Sugars 
483 
Picric acid clarification was obtained by saturating the sugar solution with 
crystals, allowing it to stand for a few minutes, filtering, and analyzing the filtrate 
immediately, so as to preclude the possibility of inversion of sucrose. 
The results show in general that clarification is unnecessary with the pi crate 
method with the materials so far analyzed. Undoubtedly the picric acid clari¬ 
fication could be effective only when proteins are present, and the latter are 
probably absent in the syrups analyzed. Therefore the above results do not 
necessarily preclude the desirability of using it in the case of other materials, 
as, for example, the vegetables analyzed by Myers and Croll {12). 
Preparations which contain chlorophyll, and which have been preserved in 
alcohol, have to be clarified with lead after dilution, since the chlorophyll pre¬ 
cipitates and must be removed. The writers suggest that, when a number of 
samples of a given material are to be analyzed with the present method, lead 
clarification be used on a few samples, and then, if no effect is noticed, that no 
clarification at all be used. 
Pig. 1 .—Color values of a series of glucose solutions from 0.02 per cent to 0.24 per cent, using the 0.08 per cent 
solution as a standard. From this graph the factors in Table IV were obtained 
CALCULATION OF RESULTS 
It is generally assumed, in using a colorimeter, that the height of the column 
of liquid required to produce a given intensity is inversely proportional to the 
concentration of the colored substance, when there is not too great a difference in 
the intensity of the two solutions. Dehn (4) has pointed out the fallacy of this. 
For example, the writers found that, if two solutions of sugars are measured, one 
just double the strength of the other, and the weaker placed at 30 mm., the strong¬ 
er reads 13.8 instead of 15. Conversely, if the stronger is set at 30, the weaker 
reads 64 mm. instead of 60 mm. The stronger is always too strong, and the 
weaker too weak. In practice, this discrepancy is met by the use of standards 
which have about the same intensities as the unknowns have. But this involves 
a series of standards. Kober ( 9) has a correction formula for nephelometry that 
applies in colorimetry also, but it was thought simpler to establish a table of 
corrections rather than to apply the formula to each calculation. The picramate 
color was prepared from glucose solutions varying in concentration from 0.02 to 
0.24 per cent, the feasible range for the colorimeter used. The procedure for 
total sugars was used; that is, the glucose was heated for 10 minutes with the 
picric acid alone, and then 30 minutes with sodium carbonate. The 0.08 per 
