( 407 ) : 
the number of umbel-rays is + 0,44, that of the number of lobes 
of the first leaf only + 0,08. To some extent this may be explained 
by the circumstance that the influence of nutrition on the first leaf 
is not so great as on characters which appear later, since the food, 
stored in the seed, is the same for both cultures and possibly has 
not been entirely used when the first leaf develops. In agreement 
with this the sensibility-coefficient of the median of the number of 
lobes of the first leaf of Scandiv Pecten- Veneris is + 0,08, whereas 
it is + 0,17 for the number of umbel-rays of the same plant. 
From what precedes it will be seen that the influence of nutrition 
of the median value of different characters of the same plant varies 
greatly, some organs being very sensitive for differences in nutrition, 
others experiencing little difference in their development on this account. 
Concerning the value of the quartile the table shows that we do 
not obtain in all the cases studied, a variation in the same sense 
by bad nutrition, as was the case with the median value. In some 
cases (J is greater in the plants from the fertile soil, in other cases 
it is smaller, as great or nearly as great as with the plants from 
» compare the variability of the 
meagre soil. In order to be able t 
characters in both cultures, however, and to draw conclusions from 
this comparison about the influence of nutrition on the degree of 
variability, we must not take the quartile but the variability-coefficient Ik 
If, to begin with, we consider the value of this variability-coefficient 
in the various cases, we see from the table that it varies between 
wide limits 0,044 and 0,36. Also VerrscnarreLrt ') found equally 
divergent values of a for the characters of different plants studied 
by him. The smallest variability is found with the different cha- 
racters of Malea vulgaris, as well in the well-fed as in the 
badly-fed plants. Hence this plant appears to be little variable. 
Comparing the variability of the different characters of the same 
species with each other, we see that they diverge relatively little 
with the well-fed plants, as well with /beris amara, as with Anethum 
graveolens and Malva vulgaris. For the different characters of Zberis 
amara 5 is respectively 0,114, 0,137, 0,18, 0,15; for Anethum 
ál 
graveolens 0,19, 0,19 and 0,18 and for Malva vulgaris 0,071, 0,089 
and 0,05. 
It will be seen that for the same species these values are nearly 
1) VerscHarreELT, |. c. p. 353. 
