56 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 
minutely rufous-pilose, with a single ovule at the base and a fleshy taper-pointed 
style at the apex; fruit (hitherto undescribed) subglobose or very broadly ovate 
and depressed at the base, with an axial length of 3.5 to 4 cm. and a diameter 
of 4 to 4.5 cm., glabrous, with the component carpels somewhat gibbous but not 
outlined by distinct areoles; seeds remarkable for their smooth polished black 
or dark brown testa, unsymmetrically obovate-oblong, often apiculate, distinctly 
marginate, and with a swollen caruncle at the base, somewhat compressed, 12 
to 16 mm. long and 8 to 9 mm. broad. (PLATE 34. Ficures 64, 65.) 
Type in the herbarium of the Royal Botanical Museum, Berlin (without 
fruit) collected near Barahona, south coast of Santo Domingo, near the Haitian 
frontier, in April, 1911, by Rev. Miguel Fuertes (no. 258). 
DISTRIBUTION: Santo Domingo. 
SPECIMENS EXAMINED: 
Santo Dominco: Duplicate of type; same locality and collector, fruits, 
September, 1912 (both U. 8. Nat. Herb.). 
LocaAL NAME: Guanabanita (Barahona, Santo Domingo). 
The closest ally of Annona bicolor is Annona globiflora Schlecht. of eastern 
Mexico. From this species it differs in its thicker, broader, coriaceous leaves, 
its smoother black-seeded fruits, and its pointed flower buds. A specimen of 
the type collection in the United States National Herbarium bearing the label 
“Anona avillifiora Spr.?” was recognized as a new species by the writer, who 
communicated the fact to Professor Urban, asking that Father Fuertes be re- 
quested to secure fruit from the type specimen, in order that the description 
of the species might be completed. Professor Urban replied that he had already 
“published the new Anona of Santo Domingo as A. bicolor in Symb. Antill. VIL. 
(June, 1912), p. 223.—Fuertes no. 258,” and added that he had not seen fruits. 
Fruits of the type plant were afterwards received by the writer directly from 
Padre Fuertes, who described them as of a green color on the exterior when 
fresh and cream color in the interior. “The fruit,” he added, “smells like 
some of the other Anonaceae and keeps the characteristic odor of the family. 
The leaves and wood are fragrant, and I do not doubt but that they would yield 
a good [aromatic] extract.” 
The species was previously collected in 1819 to 1820 by Carlos Bertero on 
the island of Hispaniola, and was referred by Sprengel to Annona awilli- 
flora DC., but the latter is a Guiana plant with much longer peduncles, as 
Professor Urban has already pointed out. From figure 64, showing a flower 
bud with one petal removed, and figure 65, showing a stamen and carpel, it 
will be seen at a glance that the present species is closely allied to the Mexican 
A. globifilora Schlecht., the type of the section Annonella. The recent discovery 
by Dr. J. N. Rose of another species of this section on the island of Santo 
Domingo, not very far from the type locality of A. UVicolor is interesting. 
EXPLANATION OF PLATE 34.—Photograph of duplicate of type in U. 8S, National Her- 
barium, together with fruit and seeds. Natural size, 
Annona rosei Safford, sp. nov. 
Section Annonella. A shrub or small tree, 2 to 6 meters high, with straight 
erect stems, slender ascending branches, and willow-like foliage; young 
branches subappressed ferrugineous pilose, at length becoming reddish brown 
or grayish, plicate-striate, bearing many conspicuous whitish lenticels; leaves 
oblong-lanceolate, the margins revolute, obtuse or rounded at the base and 
gradually tapering to a very acute apex; those at the base of branchlets fre- 
quently rounded or retuse at the apex and shorter than the succeeding ones; nor- 
