CANNA. 
Wirn this genus Linneus has, with great propriety, commenced his vegetable system; because, as that 
system is chiefly founded on the number of the anther and style, this genus, in which only one of each 
of those parts is found, necessarily occupies the first station. Even this scanty apparatus is still further simplified 
by the anther having no proper filament, but being attached, as if by accident, to the margin of one of the 
petal-like segments of the corolla, which answers the double purpose of supporting the anther, and contributing 
to the beauty of the flower; but although, to an incurious observer, this appears to be its only support, yet, 
upon a nearer inspection, we find that the margin which bears the anther, is incrassated and strengthened 
by a concealed stamen, communicating with the tube of the corolla. The style is not less remarkable for its 
simplicity than the anther; being only a long, spatulate, or rather clavate process, not tubular, but of a 
solid substance, an inch or upwards in length, and bearing at its apex a slight, and almost inconspicuous 
scale, which seems to be the true stigma ; although the upper part of the style is also frequently covered 
with the pollen from the anther. This construction of the most important parts of the flower, is so different 
from that of any other genus in the whole system, that it is impossible it should be mistaken. Those which 
approach the nearest to it, are Maranta, Phrynium, and Thalia; in all of which the anther is also attached 
to the margin of the filamentary petal, either immediately or by a short pedicel; but in the last-mentioned 
plants, the style is tubular and revolute, and the stigma an expanded orifice, the glutinous margins of which 
receive the pollen. Canna is therefore one of the most distinct and best defined genera in the whole 
vegetable system. 
But although the genus of Canna be thus readily ascertained, this is by no means the case with the 
species, the number of which now discovered is very great, and in their general habit and appearance resemble 
each other more than is observable in almost any other genus of equal extent. Thus, in all the species, the 
root is perennial, either tuberous or fibrous; the culm or stem herbacecus, and covered with the sheathing of 
the petioles; the leaves are lanceolate-acute, with too little variety either in their form or size to afford a 
satisfactory specific distinction. This, however, is found by attending to the mid-rib, which, in about one-half 
of the species, is not placed exactly in the middle of the leaf, but is nearer one side than the other, dividing 
it into two unequal lobes, and forming an ineequilateral or elliptic leaf. The inflorescence is terminal, bursting 
from a general bracte or sheath, in a loose panicle or spike; the floral bractes are wedge-shaped and deciduous; 
the flowers are generally in pairs, and open in succession. The calyx consists of three small, acute leaves, 
of nearly equal size, sitting upon the germen, and crowning the capsule when ripe. This calyx surrounds 
a tubular corolla, which expands into a double margin or limb, the outer one of which consists invariably of 
three lanceolate segments, nearly equal, an inch and a half to two inches in length, exhibiting some diversities 
both of shape and colour, in the different species; the inner border, the segments of which resemble the 
petals in polypetalous flowers, is divided into an upper and a lower lip; the upper is erect, and is generally 
divided into two or three ovate, or lanceolate segments, either intire or divided at the apex; the lower lip 
is mostly linear, and either revolute or declined. These, together with the filament and style, which rise 
between them, and which they seem intended to inclose and protect, are all united at their base into one 
tube, forming what has been considered by our elder Botanists, as an hexapetalous flower. 
It must however be observed, that in this mode of considering the organization of this genus, all Botanists 
do not agree. Thus Hegetschweiler, a German writer, published in the year 1813, a work entitled Commentatio 
Botanica, sistens descriptionem scitaminum nonnullorum, §c. in which he undertook to examine a few select plants, 
and amongst them two or three species of the genus Canna, with a view to define them with greater accuracy, 
and to trace their relations to some other genera, and particularly the orchidew. His view of the subject by 
no means corresponds with that which we have here taken; in particular, he considers that part which we 
have denominated the exterior limb or border of the corolla, as an interior calyx, and restricts the real corolla 
