VARIATIONS OF SPECIES. 355 
smaller corymbs, and smaller, thicker and pubescent leaves. 
This species is common in Virginia in dry copses and open 
woodlands, but varies greatly, so that we are puzzled in se- 
lecting typical specimens. On coming to the Piedmont 
region, however, the problem was soon solved, for here we 
found that it was no longer confined to dry and somewhat 
exposed and sterile situations, and that in proportion to the 
degree of shade or richness and dampness of soil in which it 
grew, so the leaves became thinner and larger, and the whole 
plant more robust, till it could no longer be distinguished 
from the true E. ageratoides; and on visiting the neigh- 
boring mountains, we found the latter species growing in 
great abundance. .If, therefore, the generally aecepted rule 
be applied to this case, EZ. aromaticum must be considered 
to be a variety of E. ageratoides. Ina very similar manner 
Acalypha gracilens Gray, varies into A. Virginica L., and 
it has very properly been reduced to the condition of a va- 
riety by Professor Gray. In this connection we would men- 
tion that we have found Eupatorium aromaticum with leaves 
beautifully whorled in threes. As the same arrangement has 
been observed in another species, it would seem that the 
genus is inclined towards this mode of leaf-arrangement, 
which makes that of Æ. purpureum appear less anomalous. 
Before closing we would add to the list of monccious and 
dicecious plants which have been found with androgynous 
inflorescence (see March number of the NATURALIST, p. 46) 
an instance of the same mode of inflorescence in Fraxinus 
Americana. In the spring of 1867 we observed in this 
county (Bedford Co., Va.) a tree of this species with pani- 
cles thoroughly androgynous; but in this instance, as if a 
violence had been done to nature, every flower afterwards 
became changed to a mass of small, contorted leaves, bend- 
ing the branches with their weight, and presenting a truly 
remarkable appearance. 
NOTE, — Bidens cernua and B. chrysanthemoides might also have been adduced as 
Species which run together. We beg for a sight of th tall Virgi T iu 
