A SPECIES OF PERESKIA FROM GUATEMALA. 
By J. N. Rosr. 
Since 1902 I have had knowledge of a most remarkable species of 
Pereskia in Guatemala. My attention was first called to it by Dr. 
QO. F. Cook, who collected specimens and, with Mr. G. N. Collins, 
obtained several excellent photographs. Soon after that time, and 
again in 1906, as also in 1907, Prof. W. A. Kellerman obtained photo- 
graphs and some excellent specimens. This last material, received 
coon after the death of Professor Kellerman, has led me to re-examine 
all the material at hand and to describe it as new. It has been sug- 
gested that there may be more than one species in Guatemala, but 
while my material shows considerable variation this is not such that 
I feel warranted in dividing it. This Guatemalan species has some- 
times been called P. nicoyana, but it differs from that species in habit, 
the petals are not fringed, and the axils of the leaves are not hairy. 
Pereskia autumnalis (KEichlam) Rose. Puates LII, LIII, LIV. 
Pereskiopsis autumnalis Eichlam, Monatsch. Kakteenk, 19: 22, 1909. 
Tree 6 to 9 meters high, with a large, rounded, much branched top, the trunk 
usually very definite and 40 cm, or more in diameter; young branches cherry- 
brown, smooth: axillary spines usually very slender, generally single, some- 
times in threes, 3 to 4 cm., rarely 8 cm. long; leaves thickish, oblong to orbicu- 
lar, 4 to 8 em. long by 3 to + cm. broad, round or somewhat narrowed at base, 
mucronate-tipped ; flowers sessile or nearly so; ovary bearing ovate leafy bracts ; 
sepals ovate, acute, naked in the axils; petals entire, perhaps about 1 cm. long, 
red: fruit globular, 4 to 5 cm. in diameter, fleshy, glabrous, bearing scattered 
leafy bracts, these naked in the axils: seeds black, glossy, 4 mm. long. 
The following herbarium material has been examined : 
From El Rancho, W. A. Kellerman, December 28, 1908, nos. 7011 and 7014; 
also from the same locality, O. F. Cook, April 9, 1902. 
In addition to this I have examined various bottled specimens and a fine 
series of 9 photographs, a part of which are here reproduced. 
ExpLaNATIoN oF Piatres LII, LIT, LIV.—Plates LI, LITT, two trees, showing habit. 
From photographs by W. A. Kellerman. Plate LIV, fruit and leaves. Natural size. 
aAfter this paper had gone to the printer I learned from Mr. Eichlam, of 
Juatemala City, that he was about to publish a new species of Pereskiopsis. 
His paper is now at hand and contains a very full and interesting account of 
his species. A careful reading of the description of Pereskiopsis autumnalis 
convinees me that his species is the same as the one I had proposed to publish 
here as new. I have therefore substituted his specific name in place of the one 
I had in proof. I can not agree with him, however, that it belongs to Pereski- 
opsis. The fruit and seeds are very different from those of that genus. The 
seeds of this species, as described above, accord with Pereskia while those of 
true Pereskiopsis approach those of Opuntia. 
399 
