PITTIER—REVISION OF THE GENUS INGA. 181 
SPECIES TRANSFERRED TO OR FROM DIADEMA. 
With Inga stipularis DC., placed by Bentham under Pseudinga, 
series Glabriflorae, 7. cordistipula Mart., considered as a Leptinga, 
and J. cinnamomea Spruce and J. duckei Huber, placed with Diadema, 
the Central American species /. radians Pittier and /. paterno Harms 
form a natural group, characterized by the large, foliaceous, persist- 
ent or subpersistent stipules. As the first two species further show 
the same short, capitate or clavate floral rachis which is character- 
istic of the other species of section Diadema, they should also be 
included in it. 
In his notes on Mimoseae,’ prefacing the enumeration of the species 
of Inga, Bentham states that the absolute character relied on for dis- 
tinguishing Inga from Pithecolobium is the simply pinnate leaves. 
Without contradicting this assertion I feel obliged to assume, from 
the observation of the habit and general appearance of the trees and 
from the critical examination of their floral and carpological char- 
acters, that some once-pinnate species placed in the genus Inga 
really belong to the section Caulanthon of Pithecolobium. I refer 
to Inga tubulifera, I. rufescens, I. globulifera, and I. billbergiana, 
which are very closely related to each other and at the same time 
look strikingly like Pithecolobium glomeratum, but for their once 
pinnate leaves. In Panama I collected flowering and fruiting speci- 
mens of /, rufescens and of some related forms which I take to be 
L, globulifera and I. billbergiana. The trees had unmistakably the 
appearance of the above-mentioned Pithecolobium, and this likeness 
was thoroughly confirmed by the prefloral arrangement of the buds 
and especially by the curled, dehiscent, bright red pods. Considering 
that characters drawn from the flowers and fruits should have in 
every case more weight than one single leaf peculiarity, I do not 
hesitate to propose the transfer of these species, which also should 
be reduced to two, as follows: 
Pithecolobium tubuliferum (Benth.) Pittier. 
Inga tubulifera Benth. Lond. Journ. Bot. 4: 584. 1845. 
Pithecolobium rufescens (Benth.) Pittier. 
Inga rufescens Benth. op. cit. 4: 585. 
Inga globulifera Benth. loc. cit. 
Inga billbergiana Benth. loc. cit. 
Section 3. BOURGONIA. 
CRITICAL NOTES ON SEVERAL SPECIES. 
Bentham divides this section into two groups, the one containing 
the species with axillary, elongate spikes; the other those with short 
spikes, usually clustered on defoliate nodes, nga bourgoni (Swartz) 
*Lond. Journ. Bot. 4: 578. 1845. 
36213°—16——-2 
