212 BOTANICAL GAZETTE [MARCH 
The regression straight-line formulae are: 
For ‘al oe locule and length: for 1906, /=5.689+.079 0; for 1907, 
1=6.546 
For aay ee atite and length: for 1906, /=5.948+-.542 s; for 1907, 
1=6.535+ . 299 S. 
Length of pod Length of pod 
6 7 8 9 6 7 8 10 
i fe} zt 
2 9 ; \ \ 
4 ae ; 
H ‘ 
os ic 2 —— 
6 is 
: 
' \ 
7 oe 3 Some 
8 
°o 
Bs GONE PEE, ace inh aa 
> 
Do 
i 
° 
a 
2 
Fic. 2.—Regression of pod length on Fic. 3.—Regression of pod length on number of 
number of ovules per locule; solid dots __ seeds per locule; legend as in fig. 2. 
and firm line=1906; circles and broken 
line = 1907. 
Fig. 2 shows the regression for length on ovules, and fig. 3 that 
for length on seeds. A straight line seems to represent fairly well 
the increase in pod length associated with increase in number of 
seeds. For ovules and length the agreement appears to be very 
