330 Morphology of the Fruit of Sanguinaria Canadensis 
The correlation between the length of fruiting stalk and both length of fruit 
and fertility characters of the fruit would necessarily bring about a statistical 
correlation between length of the fruit and its fertility characters, whether there is 
any other biological reason for such a correlation or not. Calculation for the 
relationship between length of fruit and its fertility for constant stem length shows 
that the removal of the influence of this latter character reduces the correlation 
but slightly. 
Clearly, then, the correlation between length and fertility of fruit is due 
mainly to causes other than the length of peduncle. 
In both series the correlation between the length of the fruit and the number 
of seeds developing is slightly but hardly significantly higher than that between 
the length of the fruit and the number of ovules formed. 
4. The correlation between the number of ovules formed and the number of 
seeds developing per placenta or per fruit is high, the constants falling immediately 
above or below -800. This is obviously a necessary condition when the ratio 
seeds/ovules is high, as it is in this case, and the quantitative statement of the 
degree of the relationship is chiefly of interest in comparison with similar constants 
from other material. It tells us nothing concerning the influence of the number of 
ovules formed per fruit upon the probability of an individual ovule developing into 
a seed. In other words, the coefficient of correlation between number of ovules 
formed and number of seeds developing is a measure of the influence of the number 
of ovules upon the absolute fertility as measured by the number of seeds developing, 
but not upon relative fertility. 
A suitable constant for the measurement of this latter relationship indicates 
that in the 1906 series as the number of ovules per fruit increases the capacity of 
the fruit for maturing its ovules into seeds decreases slightly, while in the 1907 
collection there seems to be no significant influence of the number of ovules per 
fruit upon the chances of an ovule developing into a seed. 
It has been suggested to me that the fruits in which the larger number of 
ovules are formed are the more vigorous, and that in consequence they will be able 
to develop a higher per cent, of their ovules into seeds. The discordant results 
secured in this study as well as those announced earlier for Gercis and Rohinia 
indicate that extreme caution must be used in asserting that the number of ovules 
formed in a fruit has anything to do with the proportion of them which shall 
develop into seeds. 
5. The correlations between the two placentae are all high, the direct correla- 
tion for the number of aborted ovules being actually the lowest, although the 
cross correlation for ovules of the " first " placenta and seeds developing on the 
" second " placenta is perhaps not significantly higher. The values for the direct 
correlations for seeds developing are '80 and "84 and those for ovules -89 and "92 in 
the two series. These constants show that there is a high degree of bilateral 
symmetry in the fruit of Sanguinaria. 
