40 
BOTANY: HARRIS AND SINNOTT 
Proc. N. A. S. 
bundles in the central region of the epicotyl on the other differ conspicuously 
from those found between the numbers of bundles at the two levels in the 
hypocotyl itself, in that the coefficients are extremely slight. They are 
positive in some series, but negative in others. 
The regression straight lines and the empirical means for number of 
bundles in the hypocotyl (lower line) and epicotyl (upper line) associated 
with various numbers of primary double bundles at the base of the hypo- 
cotyl in two of the lines is shown in diagram 4. The irregularity of the 
means is at least in part due to the fact that the frequencies are largely 
concentrated in the two classes of 4 and 5 primary double bundles. The 
17 
i ! 
IS 
14 
.LINE 
ft / v ft 
h / 
ts 
. Ll/VE 
98 
13 
£P/C0TY£ 
U 
(2 
/y^ — «, 
13 
II 
fi 
fi 
$2 
61 
r 
I/ 
9 
I H=- -002 +2129 P 
/ E = / 2224 + 0-033 P 
f DIMEROUS SEEM /A/ OS 
W 
DIMEROUS SEEDLINGS 
R= 1-041+ 1-993 P 
E =10-941 + 0-301 T 
1 8 
9 
4 
5 S ? 
4 
5 
6 7 8 
DIAGRAM 4 
Regression of number of bundles in central regions of hypocotyl and of epicotyl On 
number of primary double bundles at base of hypocotyl. Abscissae 
represent primary double bundles, ordinates represent mean 
number of bundles in hypocotyl and epicotyl. 
slope of the line representing change in hypocotyledonary bundles associa- 
ated with variation in number of basal bundles is steep, showing a rather 
close dependence; but the line showing the change in epicotyledonary 
bundles associated with the same variations in basal bundles is very slight, 
showing the laxness of the interdependence of bundle number in these two 
regions. 
On the basis of the data available it is impossible to assert that there is 
any relationship at all between bundle number above the cotyledons and 
bundle number below them. These results indicate that there is a com- 
plete reorganization of the vascular system at the cotyledonary node. 
