4 BULLETIN 335, U. S- DEPARTMENT OF AGRICULTURE. 
acid to form malates of potassium, calcium, or magnesium. The 
formation of these salts would still further reduce the acidity of the 
juice. 
There is a further consideration which enters into the relative acid 
content of green and ripe fruit — that of the dilution of the acidity 
by an increase of the sugar and water content. An attempt to deter- 
mine the extent to which this change affects the percentage of acid 
content was undertaken in 1912 and is treated in a section of this 
report. 
It appears, then, that there are a number of changes in the compo- 
sition of the fruit during its development and ripening which tend 
to lower the acid content. Therefore, grapes which originally con- 
tained a high total acidity might in many cases mature so as to reduce 
the final acid content of products made from them. The prime points 
of consideration for wine making are that the fruit should be well 
ripened, so that all of the combinations of the bases with the acids 
which are possible may be carried to completion and that malic acid 
may be eliminated by growth processes to the fullest extent prac- 
ticable. Then the cream of tartar in the expressed juice, whether 
it be used for unfermented fruit juice or for wine, should be pre- 
cipitated as perfectly as possible. If malic acid is present in any 
quantity in wine it will be largely changed by after-fermentation 
into lactic acid, if proper conditions are observed. 
Then a grape may be said, in a manufacturing sense, to be ready to 
harvest at that period in its development when the sugar content 
and the acid elements have reached the most favorable state (con- 
sidering the purpose for which it is intended) that is possible for 
the variety in the particular season, location, and other prevailing 
conditions. 
SOURCES AND PREPARATION OF SAMPLES. 
During the years 1911 and 1912 a certain number of plants of each 
variety investigated were reserved for the samples desired. These 
plants were selected with a view of guarding a quantity of average 
fruit so that the sampling might be done in a manner to secure typical 
material. The work of this laboratory on grape samples during the 
past four years has shown that it is not possible to select two samples 
of fruit at the same time which shall be exact duplicates. It is also 
true that one can not select samples from the same vines at different 
dates which will give results entirely logical in the sequence of the 
changing constituents. The analyses herein reported show the effect 
of these variations in sample, but these are not such as to confuse the 
results. 
The 1911 samples of Catawba, Delaware, Ives, and Norton, exam- 
ined at Sandusky, Ohio, were taken from a vineyard at Venice, 
about 3 miles west of Sandusky. The soil is heavy black loam over- 
