8 BULLETIN 1308, TJ. S. DEPARTMENT OF AGRICULTURE 
fied by 1 drop of acetic acid ; after 0.3 cubic centimeter, a dull blue, 
much brightened by acetic acid : after 0.4 cubic centimeter, a similar 
but considerably less pronounced effect ; after 0.5 cubic centimeter, a 
clear blue, apparently brightened a little by acetic acid; after 0.6 
cubic centimeter, a clear blue, not apparently affected by 1 drop of 5 
per cent acetic acid. 
The percentage of alkali combined as soap will be the difference, 
expressed as sodium hydroxide, between the percentages of total and 
of excess fixed alkali. 
THE ESTIMATION OF TOTAL PHENOLS 
The determination of the percentage of total phenols in a saponi- 
fied cresol solution is at once the most important and the most trouble- 
some of the necessary examinations. The former method (2) of the 
writer was specifically stated to be accurate only when applied to 
certain phenols, or to a balanced mixture of phenols which constitutes 
what might be provisionally termed a " normal " commercial cresol. 
As the composition of the sample in hand departs from the " nor- 
mal," so the analytical results depart from accuracy, being too low 
for low-boiling mixtures and too high for high-boiling mixtures, 
while such phenols as are difficultly volatile with steam tend to escape 
recovery. So long as the method was used in the analysis of both a 
commercial cresol and the saponified cresol solution prepared there- 
from, these discrepancies were neither very apparent nor of serious 
practical importance. But at present the method of Weiss (8) is 
largely used for analyzing commercial cresols ; and inasmuch as re- 
sults thereby obtained are not similarly affected by the boiling range 
of the sample, it is possible for the two methods to disagree seri- 
ously. Others (4) (£) have not failed to emphasize this situation 
but have offered no remedy or alternative. 
Inasmuch as the method of Weiss has received wide acceptance 
for the analysis of commercial cresol, it will be logical to extend it 
to saponified cresol solutions, if possible. The most recent de- 
scription of the Weiss method for "tar acids" is given by Hill (o) 
under the auspices of the same manufacturing firm that sponsored 
the work of Weiss. Preliminary to the application of the method 
to saponified cresol solutions it has been necessary to submit its 
possible sources of error to a more searching examination than has 
yet appeared in the literature. Hill prescribes that 25 cubic centi- 
meters of the sample of cresol shall be diluted with 75 cubic centi- 
meters of purified kerosene and the mixture distilled to decomposi- 
tion. The distillate is received in a graduated separatory funnel of 
special type, freed from water, measured, thoroughly extracted with 
a solution of sodium hydroxide, and again measured. The decrease 
in volume of the distillate is assumed to represent an equal volume 
of "tar acids." 
Two samples of commercial cresol were employed in the experi- 
ments hereinafter described — one of low-boiling range, designated 
as " Cresol L B," and one of high-boiling range, " Cresol H B." 
The distillation data of these two cresols and of the kerosene 
employed in most of the work are given in Table 1. 
One source of error in the Weiss-Hill method may lie in the re- 
moval of some kerosene along with the aqueous alkaline extract. 
