1150 
Journal of Agricultural Research 
Vol. XXX, No. 12 
American grapes grown in Ohio, New 
York, Virginia, Michigan, Pennsyl¬ 
vania, and New Jersey in the years 
1908-1913. The number of samples of 
a variety analyzed by these workers 
varied from 1 to 15 for the less impor¬ 
tant varieties to 264 for Catawba, 50 
for Clinton, 303 for Concord, 95 for 
Delaware, 125 for Ives, and 116 for 
Niagara. These were collected over 
a period of several years, the samples 
in some cases containing collections 
made in the same locality for two or 
more years in succession and in the 
case of the more widely grown varieties 
may be considered to fairly represent 
their composition as grown in the com¬ 
mercial grape-growing districts of the 
Central and Eastern States. Natur¬ 
ally no data as to the conditions of the 
various vineyards represented, or the 
climatic conditions during the period 
of the work could be obtained in an 
experiment covering so large an area, 
but it is noted (3) that in the Ohio 
grape district the year 1908 was an 
exceptionally favorable one, that 1909 
was characterized both by a very 
heavy crop and by unpropitious 
weather during the ripening period, 
and that 1910 was a year of partial 
crop failure as a result of late spring 
frosts. Hence the samples represent 
a wide range of seasonal conditions 
and of load on the vines as well as of 
soils and cultural practices and may 
be expected to show the extremes en¬ 
countered in the area represented in 
the work. The samples were in part 
taken at wine cellars as the fruit was 
delivered by wagon or rail, and in 
part shipped to the laboratory after 
having been collected by the writers, 
by cooperating members of Agricul¬ 
tural Experiment Station forces, or 
by growers, and some errors of sam¬ 
pling were inevitable. The extreme 
range in sugar content found by At¬ 
wood and his associates in the six 
leading varieties, as assembled by the 
writer from their published data, were 
as follows: Catawba—from 12.92 to 
22.53 per cent, or 9.61 per cent differ¬ 
ence; Clinton—from 11.52 to 23.24 
per cent, a range of 11.72 per cent dif¬ 
ference; Concord—from 10.41 to 20.48 
per cent, a range of 10.07 per cent 
difference; Delaware—from 16.20 to 
26.85 per cent, a -range of 10.65 per 
cent difference; Ives—from 11.17 to 
21.22 per cent, a range of 10.05 per 
cent difference; and Niagara—11.36 
to 19.24 per cent, a range of 7.88 per 
cent difference. While the maximum 
and minimum figures found for a 
variety usually occur in different years 
and in more or less widely separated 
districts, in the case of Concord they 
occurred in a group of 23 samples col¬ 
lected at North East, Pa., in 1910. 
With this exception, the samples of a 
variety for any given locality and year 
show much narrower differences, as 
would be expected. 
To sum up by a generalization, the 
range of differences in percentage of 
sugar content found by Alwood and 
his associates in samples of grapes 
collected from a large number of 
sources in six States over a five-year 
period is approximately twice as 
great as that found by the author in 
the fruit from one vineyafd over a 
like period. The absolute amounts 
of sugar present as reported by Alwood 
are uniformly higher than those found 
by the writer for the same varieties. 
This difference is due in part, undoubt¬ 
edly, to the difference in the method 
of sampling, Alwood’s samples having 
been crushed and pressed through 
cheesecloth by hand. It may in¬ 
dicate, however, that grapes grown in 
the Vineland district of New Jersey 
regularly develop a somewhat lower 
sugar content than in the Lake Erie 
and Chautauqua districts. Alwood and 
his associates (7) analyzed a number 
of samples from New Jersey in 1913, 
among them 24 varieties from the 
Vineland vineyard, and they comment 
upon them as being low in sugar as 
compared with the same varieties 
grown elsewhere. It would require a 
very comprehensive study, employing 
rigidly controlled methods of sampling, 
to definitely determine whether there 
is a constant difference in composition 
of a given variety in different localities 
throughout its range. 
PRESENCE OF CANE SUGAR 
The literature dealing with the 
chemistry of the European grape, 
Vitis vinifera, is in agreement that 
sucrose is never found in the mature 
fruit of this species. Such standard 
works on wine making as Babo and 
Mach (11), Thudichum (57), and 
Laborde (40), repeatedly make this 
statement, and the detailed studies 
of Famintzin (21), Girard and Lindet 
(23), Martinand (46), -Baragiola and 
Godet (12), and Roos and Hugues 
(54), confirm it. Martinand found in- 
vertase present in all parts of vine 
and fruit and could find no saccharose 
either in free-run juice or in the pulp 
of the ripe fruit. Girard and Lindet 
and Baragiola and Godet found traces 
of saccharose in developing grapes, 
none in mature fruit. Roos and 
Hugues studied the dextrose-levulose 
ratio in various Vinifera varieties and 
also in the American varieties Con- 
