four flat, petal-like, fomewhat unequal leaflets, in the centre 
of which isa {mall globular imperfeét coro//a with a few ftamens. 
Flowers fweet-fcented. a 
Dr. Smits appears to have had a fpecimen of this plant — 
before him when he framed his fpecific chara&ter of what hea 
fuppoled to be the radiata of Water. Butthe HypRancea 
radiata of Micnaux, probably the fame as that of Watrer, 
has entire, not lobed leaves, and flowers growing in a cyme, 
the fterile ones in the circumference, and by cultivation, ac- 
cording to Micuaux, fometimes lofes its white woollinefs on 
the under fide and ceafes to produce fterile flowers, when it 
can hardly be diftinguifhed from Hyprancea arbore/cens, 
with which this fpecies can never be confounded. 
The propriety of Dr, Smirn’s reference of the HorTENSIA 
of Jussieu to this genus is certainly ftrengthened by the con- 
fideration of this fpecies, in which three ftigmas fometimes — 
occur and the change in the fterile flowers exaétly correfpondss. _ 
the expanded calyx in both being more ufually divided into 
four petal-like, flat fegments, containing an imperfe&, globular 
corolla in the centre. The number ten appears to us the moft 
natural in the ftamens of - both, though Dr. Rota always 
obferved eleven in Hortenfia. See his obfervations on this ; 
fubje& in Aunals of Botany, vol.ii. p.g1. : 
Our drawing was made from a fine plant, fent us in flower, — 
by Mr. Loppices, in June laft. As many living fpecimens 
were brought from America laft year, by Mr. Lyons, and 
have been difperfed by his fale, this Hydrangea will foon — 
become common, fhould it fortunately be found to refit the 
cold of our winters; but being a native of Florida, where it 
was firft difcovered by Mr. Witttam Bartram, it may pro-. 
bably require the protetion of a greenhoufe. 
In its native foil it is found growing in clumps, on the 
banks of rivers; putting out many ftems from the fame root, 
five or fix feet high, and fpreading irfelf greatly on all fides, — 
by offsets. Mr. Bartram defcribes the flowers as being at 
firft of a crimfon colour, changing more purple with age, and 
finally turning brown, but not falling off. In our plant and 
in another we faw at Mr. Hissert’s, the flowers were white 
with a greenifh tinge when young, 
