366 
FREDERICK J. PRITCHAR 
^£-/G/ir o/^ /?oors /a/ ooa^c^s. 
<s:3 /^-3 /-^^ /es /as ^o.s pp.3 2es ees 30s s^t.s J6S sas ^.s -^.s ^■^s 
^ /7 a5 
\ 
/S.&5 
A 
V 
/9.e5 
9.Z93 -.03, /*i 
Fig. 6. Relationship between percentage of sugar in the beet and weight of 
root. (To accompany table II, group d,) 
The correlation between percentage and quantity^ of sugar found 
in beet roots of approximately equal weight is shown in table III a. 
Table Ill-a 
Correlation Between Percentage and Quantity of Sugar per Root, in Roots of Approxi- 
mately Equal Weight 
Nc. of 
Roots 
Weight of 
Roots in 
Grams 
Percentage of Sugar 
Quantity of Sugar 
Coefficient of 
Correlation 
Mean 
Standard De- 
viation 
Mean 
Standard De- 
viation 
371 
313 
413-447 
448-482 
483-517 
17.75^-049 
i7.7Oi.O5O 
i7.4Oi.O57 
i.49±.035 
1. 44 ±.035 
1.51^.041 
73.86=h.22I 
82.35i.252 
87.22i.289 
6.69±.i56 
7.2idz.i78 
7.59=b.205 
.99d=.ooi 
.93 ±.004 
.96^.003 
Although the weight of each group of roots covers a range of 35 
grams, the correlation between percentage and quantity of sugar per 
root is nearly perfect, varying in the calculations from .93 to .99. 
4 As the roots belonged to our breeding material and were to be used for growing 
seed they were not topped before weighing. The quantity of sugar was calculated 
on the total weight of root and crown, i. e., the total weight of root was multiplied 
by the percentage of sugar. As the crown and tail-end of the root contain a rela- 
tively low percentage of sugar the calculated quantities of sugar are somewhat 
higher than those actually present. 
