1140 
Journal of Agricultural Research 
Vol. XXX, No. 12 
also contributes to inaccuracy in 
spindle readings. Consequently such 
readings were discontinued after 1921, 
and none of the figures either for Brix 
or specific gravity are published. 
Total astringency. —Determina¬ 
tions were made by the Loewenthal- 
Proctor method, employing 25 c. c. of 
juice diluted to 750 c. c. and titrated in 
presence of indigo-carmine with N/20 
KMnO ( . Astringent nontannins were 
determined after precipitation of the 
tannins by gelatin-salt solution, shaking 
with kaolin, and filtering. The results 
are expressed in grams by the use of the 
conventional factor (1 c.c. N/20 KMn0 4 
= 0.0020785 gram tannin), although it 
is recognized that for astringent non¬ 
tannins these figures do not express 
absolute values and are servicable only 
for purposes of comparison. 
Preservation of Samples for 
Analysis 
In 1919, 1920, and 1921 the samples 
intended for analysis were duplicate 
quart bottles which were filled from 
the thoroughly stirred bulk sample, 
sealed, numbered, and pasteurized in 
the same manner as the remainder of 
the juice. The analytical work upon 
these samples had to be postponed 
until late in the autumn of 1921. In 
the meantime work upon the nature and 
amount of change in chemical com¬ 
position produced in apple juices by 
pasteurization was in progress in the 
laboratory and the results obtained 
made it seem desirable to extend the 
study to grape juices. The fact that 
sucrose had been found in considerable 
quantity in the pasteurized juice of a 
number of varieties was an additional 
reason for securing data on unpasteur¬ 
ized juices. Samples of all varieties 
were preserved in 1922 and 1923 by the 
use of bichloride of mercury. Two 
half-pint bottles were filled with the 
juice, after thorough stirring to secure 
uniformity, 0.25 gram of powdered c. p. 
bichloride of mercury was added to 
each bottle, the bottles were vigorously 
shaken to bring the mercury into solu¬ 
tion, sealed, and numbered. Two 
pint or quart bottles were filled with the 
same precautions to secure representa¬ 
tive samples, sealed, and pasteurized 
to serve for analysis of the pasteurized 
juice, as in previous years. 
Analysis of Samples of Jtjice 
Analysis of the samples of 1919-1921 
which had been shipped to Washington 
and stored at room temperature in a 
dark room, protected from freezing, 
in the laboratory, was begun late in the 
autumn of 1921. The samples were 
analyzed in the order of age, the 1919 
samples being taken first, then those 
of 1920. By the time these analyses 
were completed the work was inter¬ 
rupted by the harvest of 1922. The 
sugar determinations on the 1922 
samples preserved with bichloride of 
mercury were begun a few days after 
the receipt of the samples at the labora¬ 
tory and were completed within three 
weeks, taking the samples in the order 
in which they had been prepared so 
that none of them stood longer than 
four weeks before being handled. 
Upon completing the work with these, 
the analysis of the pasteurized samples 
of 1921 and 1922 was taken up and 
completed prior to the 1923 harvest. 
On receipt of the 1923 samples at the 
laboratory, sugar determinations were 
at once made upon those preserved 
with bichloride of mercury, followed 
by analysis of the pasteurized samples 
of the same crop. So, the pasteurized 
samples of the crop of 1919 had been 
in storage a little more than three years 
when analyzed, those of 1920 about two 
and one-half years, those of 1921 about 
fifteen months, those of 1922 and 1923 
five to seven months each, when the 
analyses were made. All samples were 
stored together, for the most part in 
the original shipping crates, in a dark 
room in which the temperatures ranged 
between 50 and 95° F. until taken out 
for analysis. 
A question which naturally arises, 
namely, whether the analytical data 
obtained from juices pasteurized and 
stored for a period of three years is 
comparable with that from samples 
stored only a few months, can be 
definitely answered in the affirmative. 
The coagulation and precipitation of 
pectins, tannins, and other constituents 
which result from the heating of a 
juice in pasteurization continues for a 
limited time after the juice has cooled 
and been placed in storage, but comes 
to an end in the course of six to twelve 
months. Repeated analyses of samples 
of the juices here used, as well as the 
results of extensive studies of the 
changes induced by pasteurization in 
both apple and grape juices to be 
reported in detail in another publi¬ 
cation, make it clear that a sterile 
pasteurized juice, stored in darkness 
and protected from freezing or over¬ 
heating, attains a constant composition 
within a few months and will remain 
unchanged indefinitely under such 
conditions { 27 ). 
