SAFFORD—CLASSIFICATION OF ANNONA, 33 
wrinkles of the endosperm; ripe pulp yellow or orange-colored, fibrous, and 
fragrant, like that of A. purpurea. (PLATES 15, 16.) 
Type in the Botanical Museum of Copenhagen, collected at Tlatatla in the 
mountains of Veracruz, Mexico, July, 1841, by Liebmann (no. 25). 
DISTRIBUTION: Mountains of Veracruz, Mexico, Guatemala, Salvador, and 
possibly Venezuela; in cultivation on the Island of Trinidad. 
SPECIMENS EXAMINED; 
GUATEMALA: Between Secanquim and Cahabon, mountains of Alta Verapaz, 
April 20, 1904, with young flower buds inclosed in involucre, Cook 87; 
Cucanhé, near Tucurt, Alta Verapaz, June 6, 1904, herbarium speci- 
mens and photographs of flowers, leaves, and young fruit, Cook 314, 
photographs 7463, 7464. (All U. 8. Nat. Herb.) 
Satvapor: Without definite locality, but presumably near the city of San 
Salvador, Renson 271. 
ISLAND OF TRINIDAD: Botanical Garden, as I. prestoei Hemsl. (Herb. John 
Donnell Smith). 
LOCAL NAMES: Soncoya, Sincuya (southern Mexico and Guatemala) ; Matacuy 
(Alta Verapaz, Guatemala). 
Annona involucrata is very closely allied to A. purpurca, of which it may pos- 
sibly prove to be a broad-leaved variety. The types of both species are Mexican. 
It is distinguished principally in the short abrupt acumination of the leaves, 
with a small terminal point which is often retuse or at least obtuse, while the 
leaves of the typical A. purpurea are more gradually acuminate. The flowers 
of the two species can scarcely be distinguished from each other. In both of 
them there is considerable variation in the size and shape of the petals, and in 
both the young flower bud is inclosed at first in an envelope of imbricating 
bracts. 
On carefully comparing a specimen of the leaves of A. prestoei Hemsl., in 
the herbarium of Capt. Donnell Smith, with the leaves of the present species, 
the author could not discover points of difference between them warrant- 
ing their separation into distinct species. A comparison of Hemsley’s figures 
above cited with the photographs presented in the present paper indicates that 
the two species are in all probability identical, Further investigation may 
show that they ought both to be referred to A, purpurea Moc, & Sessé. 
Obtuse-pointed fruits collected in Costa Rica were received from Mr. Carlos 
Wercklé, who, in answer to questions relating to the hook-pointed fruits, said 
that he had never seen fruits of that character. The fruits referred to by him 
in a paper published in the Tropenpflanzer, cited above, must have been of 
the blunt-pointed form, and consequently they are here referred to the present 
species. Subsequently Mr. Wercklé forwarded hook-pointed fruits to the United 
States Department of Agriculture, stating that in his opinion the two forms are 
variations of the same species, and in a letter dated July 15, 1913, he writes: 
“There is absolutely no difference in the flowers and the leaves between the 
soncoyas. On some fruits the pyramids are more sharply pointed than on 
others, sometimes even on the same tree, and often more or less recurved, espe- 
cially on young fruits.” It is, therefore, quite possible that Annona involucrata 
Baill. together with A. prestoci Hemsl. will eventually have to be reduced to 
A. purpurea Moc. & Sessé, and it is also possible that the type of Hemsley’s 
A. prestoei, growing in Trinidad, was introduced upon that island from Vene- 
zuela, in which case it is to be identified with A. manirote H. B. K., a species 
which also proves to be identical with A. purpurea Moc. & Sessé. 
EXPLANATION OF PLaTeEs 15, 16.—Pi1. 15, leaf, flower, gynecium, and immature fruit 
with involucre beneath the calyx, collected by Mr. O. F. Cook at Cucanhf, near Tucurt, 
Alta Verapaz, Guatemala. Natural size. Photograph by C. B. Doyle. Pl. 16, branch with 
old leaves and an extra-axillary sessile flower bud inclosed in an involucre. Natural size. 
