Few collections of three species were available when I started studying the flora
of Hispaniola: P domingensis Urban, P. depressa Urban & Ekman and P. glomerata
Urban. I have made several collections of plants in this genus, and in trying to
establish the differences between these three species and in comparing them with
the Cuban and Bahama specimens of P. bumelifolia Griseb., I came to the conclusion
that a variable species is involved that should be named P. bumelifolia
Grisebach, and with a second, little-known but distinct species, P. glomerata Urban.
In his description of P. domingensis, Urban gave the leaf-shape as the only
outstanding character separating it from P. bumelifolia. Pax in Engler, [Pflanzenreich
IV. 147. XIII(68): 1919] seems to separate the two taxa by the leaf-shape
also; at the time of Pax's studies, both species were known only from the type
collections; P. bumelifolia known only from a staminate specimen, and P domingensis
from a pistillate specimen. M y several collections show a tendency in the
Hispaniolan races to have wider, more elliptic leaves, whereas the Cuban and
Bahama specimens have more eUiptic-lanceolate leaves; some Cuban specimens
agree perfectly with the Hispaniolan form. I see no point in keeping the two
Pera depessa Urban & Ekman was described from Ekman 9564, collected on
Gonave Island, and the differences are noted as follows: P. domingensis has longer
petioles, shiny leaves, the involucres are globose-obovate and short-pedunculate;
m y collections show much variation in leaf-shape and petiole-length, and some have
dull leaves; the immature involucres in Ekman 9564 were noted as having a "short
peduncle"; according to my own observations the young involucres are nearly
sessile. The peduncle develops as the involucre grows; after the latter is shed,
the pistillate flowers begin to mature, the individual flowers develop a noticeable
pedicel, and at the end w e have a 2-3-fruited infructescence. I do not think that P.
depressa should be maintained as a separate taxon.
Pera glomerata Urban seems to be slightly different, though the fruits are
unknown; it has a glomerate staminate inflorescence and the leaves are densely
lepidote beneath. Only two collections are known, both from Tortue Island, and
more specimens, particularly with pistillate flowers and fruits, are needed for
further study. The leaf-shape agrees in general with that of P bumelifolia.