June 15,1925 Environment and Chemical Composition of Grape Juices 1155 
with 17.31 per cent sugar and 0.91 per 
cent acid. With these exceptions, the 
juices studied by Hartmann and Tol- 
man had ratios between 1:12.5 and 
1:17.5, with 1:15 as the mean. If 
Hartmann and Tolman’s results are 
fairly representative of commercial 
Concord grape juice, as their number 
and distribution would indicate, it may 
be assumed that the public finds juices 
having such a range in acid-sugar ratios 
acceptable for beverage purposes in so 
far as these constituents are con¬ 
cerned, while those aboVe or below 
these limits fall into the high acid and 
subacid groups, respectively. Of 
course, the palatability of a juice does 
not depend wholly upon the acid-sugar 
ratio, but this is a factor of outstanding 
importance. 
Carrying out the arrangement of the 
varieties into groups upon this basis, 
we have the three groups constituted 
as follows: 
High-acid group (acid-sugar ratio 1:12 or 
higher): Barry, Clinton, Clevener, Cynthiana, 
Dakota, Diogenes, Franklin, Herbemont, and 
Lenoir. 
Medium or balanced group (acid-sugar ratio 
about 1:15): Berckman, Canada, Catawba, Con¬ 
cord, Diamond, Elvira, Goethe, Isabella, Ives, 
Missouri Riesling, Nectar, Noah, Norton, and 
Worden. 
Subacid group (acid-sugar ratio 1:17.5 or less): 
Agawam, Brighton, Brilliant, Colerain, Diana, 
Delaware, Dutchess, Eumelan, Herbert, Jefferson, 
Lampasas, Lindley, Lucile, Martha, Massasoit, 
Merrimac, Montefiore, Niagara, Perkins, Pockling- 
ton, Salem, Ulster, Vergennes, Wilder, and Wood¬ 
ruff. 
Some varieties do not fit very closely 
into this classification because of varia¬ 
tions in acid-sugar ratio which are con¬ 
siderably wider than the limits of any 
one group. Examples are Agawam, 
Barry, Colerain, Diogenes, Ives, and 
Wilder. These are varieties which 
show a wide variation in acid or in 
sugar content or both, as may be seen 
by reference to the analytical data in 
Table III, with the result that the ratio 
in some years stands apart from the 
others. Such varieties have been 
placed in the groups to which the 
results for the remaining years, to¬ 
gether with the average, would place 
them. Canada, Montefiore, and Per¬ 
kins in one year (1922) had crops of 
very high acidity and minimum sugar 
with resulting ratios materially higher 
than for other years. The very unfav¬ 
orable weather conditions prevailing 
during the ripening period in 1922 
affected several other late-maturing 
varieties in the same way. The widest 
variations in ratio are found in the 
subacid group, Brighton, Brilliant, 
Eumelan, Lindley, and Wilder showing 
especially wide ranges. Two varieties, 
Noah and Norton, stand on the border 
line between the high-acid and the 
balanced groups. 
Broadly speaking, the fact that the 
acid-sugar ratio is as constant as the 
figures show it to be when the crops 
were produced under the variety of 
seasonal conditions encountered at 
Vineland in 1919-1923 shows con¬ 
clusively that knowledge as to the 
ratio for a variety is of material assist¬ 
ance in indicating its possibilities as a 
source of beverage juice. 
TOTAL ASTRINGENCY, OR TANNIN, CON¬ 
TENT 
The suitability of a fruit juice for 
beverage purposes depends not only 
upon the amounts of sugar and acid 
it contains but also upon the amounts 
of astringent constituents present. 
A palatable -beverage juice must have 
a certain fairly definite balance or 
proportion between the three constit¬ 
uents—sugar, acid, and tannin. If 
the tannin is deficient in amount the 
juice will be lacking in sprightliness 
and agreeable aftertaste; if it be in 
excess the juice will be harsh and 
astringent with a suggestion of bitter¬ 
ness. The insipid character of the 
juices of many of our choicest table 
varieties of grapes and the harshness 
and lack of palatability of the juices of 
most wild grapes are instances of these 
extremes, while Concord juice is an 
example of a fairly well balanced juice. 
The amount of astringent material 
present is as important as the sugar- 
acid ratio in determining the value of 
a grape juice for beverage purposes. 
While a considerable number of 
methods for the determination of 
total astringency and astringent non¬ 
tannins in fruit juices have been pro¬ 
posed, none of them has displaced the 
Loewenthal-Proctor method, which re¬ 
mains essentially as described by Neu- 
bauer in 1872 (50) with the difference 
that powdered kaolin is substituted for 
bone black in filtering off the gelatin- 
tannin precipitate. For quantitative 
purposes this method is faulty in 
several respects. The chief defect 
is that all substances which reduce 
KMnC >4 are included in readings as 
tannins, either wholly or in part. 
The products of the reaction of tannin 
with gelatin are variable and the pre¬ 
cipitation is a colloidal one which 
varies with conditions, as Wood (62) 
has shown. The filtration of the tan¬ 
nin-gelatin precipitate with kaolin 
involves a variable error, since the red 
pigment of some deeply colored juices 
is very strongly adsorbed by the filter 
paper used and can not be wholly re- 
