494 Harris: FACTORS INFLUENCING THE,WEIGHT OF BEAN SEED 
For pods and seed weight, o¢, = .0916 + .0084. 
The differences in the variabilities of the three sets of corre- 
lations are therefore: 
For pods and ovules and pods and seeds, .0047 + .O117. 
For pods and ovules and pods and weight, .0037 + .O118. 
For pods and seeds and pods and weight, .oo10 + .O1i8. 
All of these differences are only a fraction of their probable 
errors. 
It is quite clear from the irregularity with which the values of 
the correlations for pods and ovules and for pods and seeds are 
scattered about the line showing the distribution of the correlations 
for number of pods per plant and seed weight, that there is very 
little if any interdependence between these three correlations. 
IV. RECAPITULATION 
This paper, which is one of a series dealing with the physiology 
of seed production, presents constants showing the degree of cor- 
relation between the weight attained by the individual seed and 
the number of pods per plant. 
The correlations are positive throughout the twenty-seven 
series for which data are available, but are of a low order of mag- 
nitude and highly variable. They range from +.005 to +.339) 
with a mean of +.159 + .012 and an absolute variability of 
-092 + .008. 
In both average magnitude and variability the correlations for 
seed weight and number of pods per plant are in excellent agree- 
ment with those for number of ovules per pod and number of pods 
per plant and with the correlation for number of seeds per pod and 
number of pods per plant. 
The average value of the correlation iss pods and weight is 
lower than that for pods per plant and ovules per pod and higher 
than that for pods per plant and seeds per pod, but both of these 
differences are low, and may not be significant in comparison with 
their probable errors. 
These studies will be continued. 
CoLp Sprinc Harsor, NEw York. 
