PEOCESS OF ftlPEttlttG I£f THE TOMATO. 15 
METHODS OF ANALYSIS. 
Sampling and preservation. — In order to ootain representative 
samples at each stage of ripening and to avoid the necessity of 
analyzing a large number of individual fruits to determine existing 
variations, composite samples were resorted to. These composite sam- 
ples were taken from approximately 20 tomatoes. To eliminate error 
due to possible correlations between size and chemical composition, 
tomatoes were chosen so that each composite sample was obtained from 
fruits of all sizes, with the exception of abnormally large or small 
fruit which were discarded. The method of collecting the samples 
was uniform throughout. Where the fruits were small (e. g., those 
14 days old) a 200-gram lot was made by using entire tomatoes, but 
with larger fruit samples of 200 grams were made up by removing a 
cylinder from each tomato with a half -inch cork borer. The cylinders 
were taken through the equator perpendicularly to the axis. A fairly 
representative sample was obtained in this manner, for the portion 
removed from each tomato was roughly proportional to the size of 
the whole fruit. The method of preserving samples for analysis was 
similar to that used by Hasselbring and Hawkins (21) in their studies 
of sweet potatoes and identical with the procedure of Kraus and 
Kraybill (28) with tomatoes. The material was heated with 80 per 
cent alcohol for 1 hour at 70° to 75° C, with the addition of cal- 
cium carbonate (CaC0 3 ) to insure the neutralization of acids. Two- 
quart glass-top jars were used, and approximately 1,065 c. c. of 95 
per cent alcohol and 0.5 gram of precipitated CaC0 3 were added, 
after which the heating was carried out on a boiling water bath. 
Moisture and ash samples were merely covered with 95 per cent 
alcohol without subsequent heating. 
In preparing the samples for analysis (with the exception of certain 
moisture, dry-weight, and ash samples) the alcohol was removed from 
the insoluble residue by filtering into a 2-liter volumetric flask. The 
residue was thoroughly extracted with warm 80 per cent alcohol, 
which was cooled, filtered, and added to the original filtrate. The 
volume of the flask was then made up to mark at 20° C. (referred 
to later as the original extract) and one-tenth and three-twentieths 
aliquots pipetted off and placed in separate Florence flasks, which 
were stoppered, labeled, and set aside. The residue was dried at 
80° C. in an air oven for a few days and then allowed to come to 
air-dry weight, after which it was weighed and finely ground in a 
drug jnill (referred to later as the original residue) . One-tenth and 
three-twentieths portions were weighed and stored in small stop- 
pered vials. 
Moisture, dry weight, and ash. — An entire 200-gram sample covered 
with 95 per cent alcohol was placed in a large beaker and evaporated 
nearly to dryness on a steam bath. It was then transferred to a 
