ht-cap, on the falling off of which the
whole of the capitulum is perceived to be of a green colour, shortly the
crown of it becomes of a fine reddish purple, this colour extends itself
gradually downwards, presently we see the upper half of the head purple,
the lower half green, in this state it has a most pleasing appearance;
the purple still extending downwards, the whole head finally becomes
uniformly so, and then its flowers begin to open, and emit an odour
rather agreeable than otherwise; on dissecting a flower we find three of
the stamina in each longer than the others, and bearing two little
points, which proceed not from the antherae, but from the top of the
filaments, it is therefore one of those Alliums which LINNAEUS describes,
as having _Antherae bicornes_.

This species increases readily by offsets, which should be separated and
planted in Autumn.

We know not why LINNAEUS should give it the name of _descendens_, unless
from its being one of those plants whose roots in process of time
descend deeply into the earth.

[Illustration: _No 251_]

[Illustration: _No 252_]

[252]

CAMPANULA GRANDIFLORA. GREAT-FLOWERED BELL-FLOWER.

_Class and Order._

PENTANDRIA MONOGYNIA.

_Generic Character._

_Cor._ campanulata fundo clauso valvis staminiferis. _Stigma_ trifidium.
_Caps._ insera poris lateralibus dehiscens.

_Specific Character and Synonyms._

CAMPANULA _grandiflora_ caule subunifloro, foliis sublanceolatis
serratis, corolla patente. _Jacq. in Litt. Hort. v. 3. t. 2._

CAMPANULA _grandiflora_ foliis ternis oblongis serratis, caule unifloro,
flore patulo. _Linn. Suppl. p. 140. Syst. Veget. ed. 14._ _Murr. p.
207._ _Ait. Kew. v. 1. p. 218._

Professor JACQUIN is, we believe, the first author who has figured this
species of Campanula, which he has done in his _Hortus Vindebonensis_;
LINNAEUS the Son afterwards inserted it in his _Suppl. Pl._ assigning it
the characters specified above in the synonyms, and expressing his
doubts whether it was not a variety of the _Campanula carpatica_,
already figured in this work, _Pl. 117_. Prof. JACQUIN clearly
demonstrates that it cannot be so, as it differs most essentially from
that plant in a variety of particulars, _vid. Linn. Syst. Veget. ed. 14.
Murr._ his specific description there given, agrees much better with the
plants we have seen flower here, than that of LINNAEUS does, there being
generally more than one flower on a st