112 
CONTRIBUTIONS TO BOTANY. 
On the CanellacejE. 
Having been engaged for a long time in the study of the Clu- 
siacece, 1 have been led to examine the several genera that have 
been referred at different periods to that family, and in this 
manner Canella and Cinnamodendron came under my especial 
notice. The former genus was arranged by DeCandolle and 
Choisy in Guttifera, but it was partially separated from the 
order by Martius and Endlicher, and made the type of a sub- 
order of that family under the name of the Canellacea, asso- 
ciating with it Cinnamodendron and Platonia. It will, however, 
be seen that Canella has no relation with Platonia, but that its 
real affinity tends towards the Winteracece. I will therefore 
proceed to state my reasons for this conclusion, will expose the 
characters of both groups, and describe their genera severally. 
This small family consists only of the two genera above men- 
tioned, Canella and Cinnamodendron. They form evergreen 
trees, with a bark possessing the taste and smell of cinnamon ; 
they have a copious foliage of alternate, somewhat fleshy, exsti- 
pulate leaves, which ai’e furnished with dotted glands, and have a 
taste similar to that of the bark : the flowers are small, in short 
axillary or terminal corymbs, having a persistent calyx of 3 sepals, 
5 petals with dotted glands, or sometimes 10 petals in 2 series, 
extrorse monadelphous stamens, a unilocular ovarium with two 
or more parietal placentations ; a small, baccate, 1-celled fruit, 
containing a few black, shining, reniform seeds having a parietal 
attachment. Many of these characters are possessed by the 
Winteracea;, from which the Canellacece differ in the union of 
their stamens into a monadelphous tube ; in the shape and dis- 
position of the anther-cells •, and in lieu of several distinct car- 
pels, they have only one solitary unilocular ovary, with 2 to 5 
parietal placentations. In the former family, whether there be 
sevei’al ovaria, or whether by abortion they be reduced to one, 
these, although 1-celled, invariably exhibit a single placentation 
along the ventral face. 
1. Canella. 
The plant upon which this genus was founded was first de- 
scribed by P. Brown in his ‘ History of Jamaica.'’ During a long 
time it was considered to belong to the same genus as Drimys 
Wintei'i, a small tree growing in Tierra del Fuego, near the 
Strait of Magellan ; for the bark of both species had for some 
time been used in commerce, as mentioned by Clusius in 1605, 
under the name of Cortex Winterana, and confounded with the 
white bark Canella alba, a name afterwards given exclusively to 
the Jamaica kind. By Linnseus the latter was also considered 
