at the top into three petioles bearing five lanceolate leaflets, 
fawed at the edges, fhining underneath, nearly feflile. Peduncle 
from the bofom of ‘the petioles, ereét, longer than thefe and 
equal to the whole leaf. Flowers in a clofe hemifpherical 
umbel, with a many-leaved involucre the length of the pedicles, 
which are fhort. Calvx fuperior, very obfoletely five-toothed. 
Petals white, five. Ssamens five, the length of the petals, 
ofien wanting. Germen three-cornered, three-celled. Svy/es 
three, Stigmas obtufe. We have obferved no planis bearing 
male flowers only, but many of the florets were female in moft 
of the umbels that we examined, and in the one from which 
our drawing was taken, all of them; but whether originally fo 
or only from the falling of the ftamens we are uncertain. 
In variety (8) of which we have only feen dried {pecimens, 
the leaflets are fhorter, more oval, and rounded at the bafe, 
and are well reprefented in PLUKENET’S figure above quoted, 
It is not improbable that our two varieties may be hereafter 
confidered as diffinct fpecies; and for this reafon we have 
thou ht it belt to limit the name of ¢rifolia to the latter, to 
which alone it is at all applicable; for aon variety (a) 
may occafionally produce only four or three aflets, ag in 
Enrer’s figure, yet this is evidently from abortion; which 
does not appear to be the cafe in (8). 
There is no ground whatever for Linn aus’s fufpicion that 
this plant might be the male of Panax quinquefolia. No two 
{fpecies can be more diflin@. The three diftin@ fiyles would 
' rather lead to the idea of a different genus: the fruit we have 
never feen. 
Native of Penfylvania, Maryland, New-York, and Vir- 
ginia. Our drawing was taken from a plant communicated by 
Melfirs. Loppicgzs and Sons; but we defcribed it from 
fpechmens in Mr, Vsre’s garden at Kenfington-Gore, in 
May 1807. 
