PITTIER—REVISION OF THE GENUS INGA, 217 
this and J. vera typica occur here and there. The facts do not warrant the 
recognition of this form as a distinct species. 
Inga vera portoricensis Pittier, subsp. nov. PLaTE 104. 
Flowers of the typical plant, but the leaflets seeming stiffer and more cori- 
aceous, with the costa and veins thicker and very prominent on the lower face; 
pubescence of the young growth, the rachis of the leaves, and the peduncles 
ferruginous, but the density of the indument variable; glands varying in size 
and shape and often obsolete. 
Type in the U. S. National Herbarium, no. 362665, collected 40 miles northeast 
of Mayagiiez, Porto Rico, by A. A. Heller (no. 4471). 
Apparently endemic to Porto Rico. 
In conclusion, it seems that the typical Inga vera of Willdenow is an ex- 
clusively Antillean species, restricted to the islands of Haiti and Jamaica, and 
that aberrant forms, the result of isolation, have developed within the area of 
the species in Cuba, Porto Rico, Haiti, Jamaica, and perhaps Trinidad. These 
forms, however, have not reached a stage of differentiation sufficiently marked 
to justify their being considered as distinct species. It is to be noted, more- 
over, that the 30 specimens at hand all proceed from the above-named islands, 
to the exclusion of the Lesser Antilles. 
EXPLANATION OF PLATE 104.—From a field photograph of Inga vera portoricensis taken 
at Cayey, Porto Rico, in 1899, by Mr. G. N. Collins. Natural size. 
GROUP OF INGA EDULIS. 
In this group the calyx is long and narrow or short and broad. In the first 
case, illustrated in I. edulis, the flowers are sessile, while in the latter case, 
represented by I. ingoides (illustrated in pl. 105) and I. subnuda, they are 
long-pedicellate. This last character excludes numerous specimens of Central 
America thus named in our collections. So far as is known, and excepting a 
specimen of doubtful identification from Bolivia (Williams 661, in the New York 
Botanical Garden Herbarium), Inga ingoides is restricted to the Windward 
Islands, Trinidad, and the adjacent coastal parts of Venezuela and the 
Guianas. 
Inga edulis has a form with a short calyx. Further, this species, I. oer- 
stediana Benth., and I. eriorhachis insensibly grade into each other in such a 
way as often to make the identification difficult. The two latter species seem 
to be simply high altitudinal variations of the former. 
EXPLANATION OF PLATE 105.—Two fruits of a Guadeloupe specimen of Inga ingoides 
Willd., in U. 8. National Herbarium, collected by Pére Duss (no. 3035). Natural size. 
GROUP OF INGA SPURIA. 
To judge by the existing confusion in the collections consulted, the species 
of this group, and especially the Central American ones, are very difficult to 
recognize. Bentham himself seems to have been in doubt more than once as 
to the standing of some of them, and in his last work on the genus he reduced 
to synonymy several of his own creating or the work of other authors, two 
of which, at least, appear to Have been well founded. 
Inga spuria does not seem to be quite the same thing at both extremes of 
its vast area of dispersion. Taking the Venezuelan specimens as typical, the 
species having been described originally from Carfipano, we find, for instance, 
that they possess a broad, short, stipitate calyx, with triangular, acute, rather 
short teeth, In Mexican specimens the same part may be narrower and 
