A TRIBE OF LEGUMINOS2. 25 
in Geoffroya the flowers are yellow and feetid, in simple racemes. 
To this may be added that the calyx is broader, less oblique, and 
less deeply toothed in Andira than in Geoffroya, the former ap- 
proaching more to the habit of Macherium, the latter to that of 
Pterocarpus. The arrangement of the leaflets is variable in both 
genera. 
The species of Andira present in many instances considerable 
difficulty. After deducting the А. Amazonum, whose characters, 
if confirmed by the fruit, may possibly be considered as generie, 
and the well-marked А. cujabensis and A. cubensis, the remainder, 
although widely different in extreme cases, seem to pass so gra- 
dually into each other that their delimitation seems more uncer- 
tain than in the case of any other Dalbergiee; and although I 
have now considerably reduced the species which I had originally 
established from almost solitary specimens, it is not unlikely that, 
When better known, some of those I have still retained may prove 
to be varieties only. The Geoffroyas appear to be better defined, 
but, with the exception of the widely-spread G. superba, they 
are as yet scarcely known but from single specimens. 
Dipreryx and Preropon. 
This is another pair of closely allied genera, forming a little 
group readily distinguished among Leguminose by the form of the 
calyx and petals. The petal and stamen-bearing disk lines the 
Whole of the short tube of the calyx, of which the two upper lobes, 
large and oblong, are closely pressed against each other, face to 
face in the bud, and entirely enclose the petals, the minute lower 
lobes forming a scarcely perceptible lower lip, resembling in the 
bud a small outer scale at the base of the upper lobes. The wings x 
of the corolla are notched or bifid, which does not occur, to шу 
knowledge, in any other Papilionacea, and the only approach to the 
above-mentioned structure in the calyx is in Mr. Spruce’s new 
genus Monopterye among Sophoree, where, however, the two 
upper lobes are united on the upper side to near the apex. The 
structure of the ovary and the drupaceous legume place these two 
genera in the same group as Andira and Geoffroya. In their 
foliage, the usual deficiency of the terminal leaflet distinguishes 
them from all other Dalbergiee, but this character does not appear 
to be quite constant. Between the two genera the chief differ- 
ences are in habit, the leaflets being few, large, and coriaceous ш 
Dipterys, more numerous, smaller, and thinner in Pterodon ; in 
the consistency of the upper lobes of the calyx—coriaceous 1n the 
