COLLECTED IN HONG-KONG. 319 
genera would seem to be identical ; but so far as my materials allow 
me to judge, I am inclined to think that Sedgwickia should be kept 
separate. How far the other two may be distinct I have not the means 
of determining ; but that they should all range along with Bucklandia 
in Zamamelidacee I have no doubt. Sedgwickia is the only one of the 
gie of which I possess specimens, and they are.from Griffith himself. 
In the flowers are rather too mature to be examined with 
accuracy, but they seem to be hermaphrodite, as indicated by Griffith. 
The flowers forin a dense globose head, and are all connate, that is, the 
calyees adhere to each other by their external walls, and by their inner 
to the ovary, to within a little of its apex. The limb is divided into 
about twelve short irregular, somewhat triangular, lobes, between which 
and the styles the stamens are inserted on an epigynous disk. The 
stamens are numerous, perhaps about twenty, but I have not been able 
to make out the exact number. e filaments are so very short that 
the anthers may be said to be almost sessile, and these latter are oblong, 
turned inwards, two-celled, with the cells opening by longitudinal slits. 
he ovary is more than half-inferior, and two-celled. e ovules are 
very numerous, roundish, flattened, and attached to placentz which 
adhere to the dissepiment. The styles are subulate, revolute, and 
grooved, and stigmatose on their inner face. 
What Griffith's reasons were for supposing that Sedgwickia was iden- 
tical with Altingia, I have no means of ascertaining; but that such a 
conclusion was premature, is, I think, certain, if the characters of 
the one genus be compared with those of the other. Tt was, however, 
this assertion, together with some remarks on the relations of the 
present Order, that first called my attention to the affinity which exists 
between Hamamelidacee and Liquidambar. Griffith’s words are— 
“The family Balsamiflue appear to be generally allied to Platanee, 
Salicinee, and some of their neighbours. And although the structure 
of Bucklandia was not detailed before 1836, it still appears to me odd 
that no indication of the similarity of Ziguidambar with Fothergilla had 
been noticed." 
Although Liquidambar is moncecious, that is not a sufficient reason 
for excluding it from Hamamelidacee, for Bucklandia is polygamous, 
and Sedgwickia may prove so likewise. Liquidambar has also the 
copious albumen of Bucklandia, which is wanting in those Orders with 
which Lindley has associated it in his * Vegetable Kingdom.” 
