Harris: FACTORS INFLUENCING THE WEIGHT OF BEAN SEED 487 
TABLE I 
z Number of |Number of Seeds} Correlation Pods 
Series Plants Weighed and Weight 7|£, 
185 2861 i572.018 13.08 
LL II4I 3947 074+.011 6.73 
GG 747 8364 055 +.007 7.86 
GGH 583 2828 173.012 14.42 
GGH2 499 1284 o8r+.019 6 
GG 396 1329 039 +.019 2.05 
D 514 2140 -I150+.014 10.71 
GGD2 I419 -240 +.017 14.12 
GGDD 342 1093 -193 +.020 9.65 
160 5778 145.0 16.11 
HH 1484 7325 -IQI +.008 23.87 
HHH r27r 5601 -264+.008 33.00 
HD 1416 6630 303 +.008 37.87 
HDD 1204 5029 -I17I+.0 19.00 
2362 031 +.014 22r 
DD 513 2362 -339 +.012 28.25 
DDD 459 1946 .077 £.015 5-13 
DH 670 3222 .221 4.011 20. 
DHH 565 2433 .307 +.012 25.58 
361 1164 -154+.019 
USHH 224 -108 +,029 3-73 
312 1002 282 +.020 14.10 
USDD 237 789 .210 +.023 9.13 
476 S1i7 .005 +&.015 
FSHH 429 1788 .053 +.016 3-31 
428 1989 205 +.015 13.66 
__ FSDD 387 1643 .069 +.017 4.06 
expressed graphically on the same diagram, as in Fics. 1 and 2,* 
the results seem very irregular indeed. This lack of uniformity 
appears in both the varying slope of the lines and in the scatter of 
the observed means about these lines. 
The variation in the slope of the lines must be considered in 
relation to the great diversity in environmental conditions under 
which the plants were grown. 
The slope of the line is determined solely by the magnitude of 
the absolute variation constants for seed weight and number of 
pods per plant and the correlation between them. The standard 
deviation of seed weight is relatively little influenced by the differ- 
ences in environmental conditions obtaining between the cultures 
here considered. The standard deviation of number of pods per 
plant varies greatly from experiment to experiment. Thus with 
the same degree of correlation, the actual rate of change in seed 
* in order to accousmmadate | as large a series as possible of figures on the same 
diagram the marginal scales have been given no numerical values. The numbers of 
pods and the end points of the lines drawn are given in the table. 
