54 BOTANICAL GAZETTE [JANUARY 
approximately the same with respect to brix and sugar, but P.E. 
sing. for acid and for ratio is considerably different in the two lots. 
Data on lemons 
That different lots of fruit show different values for P.E. sing. is 
also apparent from the analysis of individual lemons. In table VIT 
will be found the results of the analysis of thirty lemons from 
two different lemon trees, each tree representing a different strain 
TABLE V 
SHOWING DIFFERENT VALUES OF P.E. SING. WITH DIFFERENT LOTS OF ORANGES 
D Sol. sol... 
No. oF EGREES BRIX PERCENTAGE ACID acid Tatio 
TREE NO, FRUITS IN 
SAMPLE 
Mean |P.E.sing.| Mean _ | P.E. sing. Mean _ | P.E. sing. 
PRES See 12 13.70 0.5 0.87 0.07 15.9 ri 
yep iba nn Gory 13 I5.00 0.5 0.86 0.04 17.4 0.8 
£2 Oa, 12 11.80 0.4 0.79 0.08 et 3 
Se OL Cae he 9 £2.45 0.3 1.46 °.10 8.6 O.7 
TABLE VI 
COMPARISON OF COMPOSITION OF FRUIT FROM TWO GRAPEFRUIT TREES 
Bere PERCENTAGE PERCENTAGE SoLmps-AciD 
Torat SUGAR ACID RATIO 
TREE NO. NO. OF 
FRUITS 
P.E, PE: P.E. P.E 
Mean sing. Mean sing. Mean sing. Mean sing. 
Pe a SOF 33-751. 0735 | S610 [0:27 | 2. 20-10. 07 5:8} -o.2 
Be varus ye 50 | 22.30] 0.35 1.9.80 | 6.20 |-1.05 | 0:09.) 9.5.1. 0748 
of the Eureka variety. While too much reliance cannot be placed 
on the values obtained by analyzing fifteen fruits, it is seen from 
the table that the two lots of fruit probably have different values 
of P.E. sing. with respect to three of the characters of which 
analytical results were obtained. 
Further precautions regarding use of formulas 
Two further precautions may now be added regarding the use 
of the formulas. When the value of P.E. sing. has been found for 
one tree or lot of fruit, it must not be assumed that another tree 
