378 FLORA NIGRITIANA. 
tribe to which it gives its name, and the hardening of the 
endocarp, which distinguishes the polypyrenous from the multi- 
locular berry, is but a vague character. It might be better 
to suppress the two last tribes (Zsertiee and Hameliee), and 
re-distribute the whole into three: Gardeniee, with fleshy 
indehiscent fruits; Rondeletie, with dehiscent or pluricoccous 
fruits and interpetiolar stipules; and Hedyotee, with dehiscent 
or pluricoccous fruits and conpetiolar stipules. For the subdivision 
of Gardeniee, the placentation appears to afford good characters, 
although I have not as yet examined with this view the whole of the 
genera. Probably three distinct forms will be found: Eugar- 
deniee, with parietal placente not reaching the axis; Randice, 
with the ovules more or less immersed in thick fleshy placentz; 
and Bertieree, (or say Hameliee ?) with thinner placenta, 
superficial ovules, and little or no pulp to the berry. Some 
genera of the last group come into close connection with some 
Rondeletiee, and from the latter the passage is very gradual 
. into Hedyotee ; yet I am unable to suggest any better distinction 
between them than those generally adopted. Of the tribe 
Isertiee, DC., Metabolos would take its place among Hedyotee; 
next to Hedyotis (Euhedyotis, Arn.), from which it differs but 
slightly as a genus; Gonzalea among Rondeletice, close to 
Lerchea; Isertia and Bruinsmania among Gardeniee (Ber 
tieree). The greater part of the Hameliee would also range m 
the last-mentioned subtribe; fof which, on that account, the 
name Hameliee might be retained. Morelia, however, 85 
well as Alibertia, Schradera, and perhaps one or two more, 
would be classed in the subtribe Randiee. To the same sub- 
tribe I should refer Cordiera, and a few imperfectly known 
genera allied to it, in which the ovules are said to be large 
fleshy and peltate, but which have most probably large 
fleshy peltate placenta, with one or more ovules imm 
therein, but not easily distinguishable in dried specimens. 
Among the solitary-ovulated tribes, the only alteration of im- 
portance which suggests itself is the consolidation of the tw? 
tribes of Guettardiee and Coffeec into one, as neither the two- 
