864 
Journal of Agricultural Research 
Vol. XXI, No. II 
We may digress momentarily to see whether the pruning increases 
the dominance of the laterals produced in the distal region of the mother 
shoot. For example, in Table I it will be seen that the class of mother 
shoots having 23 original buds produced a mean total shoot growth of 
242.5 cm. and that the five upper buds produced 90.7 per cent of the 
total shoot wood. By comparing this with unpruned mother shoots 
having 24 original buds (Table II), it will appear that the five upper 
buds produced only 78.8 per cent of the total shoot growth. It will 
also be noted that 21 laterals were produced on the average pruned 
mother shoot mentioned, and 25 laterals were produced on the average 
unpruned mother shoot selected as an example. 
A consideration of the relative length of the laterals on the mother 
shoots prompts one to determine the nature and amount of correlation 
existing between the laterals produced near the distal end of the mother 
shoots. The results of such an inquiry might be expected to throw some 
light upon two important questions—(1) “Is there a group of factors 
which tends to differentiate the mother shoots in their capacity for the 
production of laterals ?” If so, they should find expression in a correla¬ 
tion between any two laterals on the same mother shoot. (2) “ Are there 
factors which tend to differentiate the laterals on the same axis?” If so, 
they should find expression in differences in the correlation between 
laterals in different positions. 
The coefficients given in Table VI show the correlation existing between 
the first four laterals on the pruned mother shoots when compared in 
different ways. The coefficients of ordinary, or gross, correlations are 
given at the head of each column. Below them stand the coefficients of 
partial correlation. 
The latter coefficients are intended to express adequately the amount 
of correlation existing between nonadjacent shoots. They are an attempt 
to answer such a question as, “What is the correlation between the first 
and third laterals, assuming that all the second laterals were of equal 
length ?” 
Table VI .—Coefficients of gross and partial correlation of the length of the four distal 
laterals on pruned mother shoots 
'l2 • ..O. 346 ±0.036 
'l2.8. 266 ± .038 
'12.4. 3 2 4 ± -°37 
' 12 . 34 - • .. 2 5 8 ± -038 
'13. 
. . . . O. 296^0. O37 
'l3.2. 
.2II± 
•039 
'18.4. 
. 267± 
.038 
'l3.24 • • • • 
•039 
*14 .o. i 57 dbo. 040 
*14.2. 090± . 04I 
'14.3. o8 3 ± -041 
'14.23.046 ± .041 
'23.0.313^0.037 
'23.1. 2 35 ± • °39 
'23.4... 2 7 °± • ° 3 8 
'23.14.202 ± .039 
o. 2i6±o. 039 
24.1 .I 75 ± -040 
24.3.143 ± .040 
'24.13.126 ± .040 
34.0.^721110.030 
r 31.1. 2 39 ± -039 
'34.2. 22 I i .O39 
'34.12.2 07 ± .O39 
The coefficients of gross correlation are all positive and of sufficient 
magnitude in comparison with their probable errors to be considered 
significant. They indicate that the growth of one lateral is correlated 
