428 BOTANICAL GAZETTE [DECEMBER 
its all being grown in one locality under similar environmental 
conditions, and thoroughly mixed before being used, to enhance 
its uniformity. 
Investigation 
CONTROL OF CROSS-FERTILIZATION.—Undesirable cross-fertili- 
zation was avoided by planting the seed-beets of approximately 
equal qualities in isolated groups, a practice commonly followed in 
beet breeding. At Fairfield the quality was determined from the 
percentage of sugar, but at Madison a somewhat different method 
of classifying the roots was employed. As small beets are richer 
on the average than large beets, the regression coefficient was 
determined between weight and percentage of sugar and a correc- 
tion made for size. Roots of equivalent value, as determined from 
both size and percentage of sugar, were then planted together at 
distances of 40-80 rods from other groups. However, at Brookings 
all the selected seed-beets of each variety were planted in a group 
and allowed to cross-fertilize with one another but not with beets 
of another group. 
CHEMICAL CONTROL.—AII the chemical work was done by experi- 
enced sugar chemists, the personnel consisting of JosrpH F. REED 
at Fairfield, Guy YouncBERG at Brookings, and W. B. CLARK and 
assistants at Madison. 
The sugar determinations of beet families were made at Fair- 
field by collecting roots at regular intervals in the row, grinding 
them to a fine pulp, and making composite analyses of the samples. 
A somewhat different method of procedure was followed at Madison. 
The quantity of sugar was first determined for each individual 
root and the sum of the quantities was then divided by the total 
weight of roots. 
FIELD CONDITIONS UNDER WHICH THE MATERIAL WAS GROWN.— 
The experimental fields were fairly uniform with respect to soil- 
type, fertility, drainage, and previous cropping, and all field opera- 
tions, as plowing, planting, thinning, cultivating, etc., were always 
performed in as short a time as circumstances woubl permit, in 
order to prevent the introduction of unequal factors. 
SELECTION OF FOUNDATION sTOCK.—The choosing of a variety 
as foundation stock often has as much to do with the improvement 
